id
large_stringlengths 9
16
| submitter
large_stringlengths 3
64
⌀ | authors
large_stringlengths 5
20.2k
| title
large_stringlengths 5
245
| comments
large_stringlengths 1
482
⌀ | journal-ref
large_stringlengths 4
382
⌀ | doi
large_stringlengths 9
151
⌀ | report-no
large_stringlengths 2
509
⌀ | categories
large_stringlengths 5
78
| license
large_stringclasses 9
values | abstract
large_stringlengths 83
3.55k
| versions
large listlengths 1
21
| update_date
timestamp[ms]date 2007-05-23 00:00:00
2025-05-30 00:00:00
| authors_parsed
large listlengths 1
1.35k
| prompt
large_stringlengths 166
3.64k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2207.03579 | Xuanwen Huang | Xuanwen Huang, Yang Yang, Yang Wang, Chunping Wang, Zhisheng Zhang,
Jiarong Xu, Lei Chen, Michalis Vazirgiannis | DGraph: A Large-Scale Financial Dataset for Graph Anomaly Detection | Accepted to NeurIPS 2022. Dataset Url: https://dgraph.xinye.com/ | null | null | null | cs.SI cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Anomaly Detection (GAD) has recently become a hot research spot due to
its practicability and theoretical value. Since GAD emphasizes the application
and the rarity of anomalous samples, enriching the varieties of its datasets is
fundamental work. Thus, this paper present DGraph, a real-world dynamic graph
in the finance domain. DGraph overcomes many limitations of current GAD
datasets. It contains about 3M nodes, 4M dynamic edges, and 1M ground-truth
nodes. We provide a comprehensive observation of DGraph, revealing that
anomalous nodes and normal nodes generally have different structures, neighbor
distribution, and temporal dynamics. Moreover, it suggests that unlabeled nodes
are also essential for detecting fraudsters. Furthermore, we conduct extensive
experiments on DGraph. Observation and experiments demonstrate that DGraph is
propulsive to advance GAD research and enable in-depth exploration of anomalous
nodes.
| [
{
"version": "v1",
"created": "Thu, 30 Jun 2022 07:16:03 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Jul 2022 02:53:29 GMT"
},
{
"version": "v3",
"created": "Thu, 25 May 2023 09:00:07 GMT"
},
{
"version": "v4",
"created": "Fri, 9 Jun 2023 11:37:55 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Huang",
"Xuanwen",
""
],
[
"Yang",
"Yang",
""
],
[
"Wang",
"Yang",
""
],
[
"Wang",
"Chunping",
""
],
[
"Zhang",
"Zhisheng",
""
],
[
"Xu",
"Jiarong",
""
],
[
"Chen",
"Lei",
""
],
[
"Vazirgiannis",
"Michalis",
""
]
] | TITLE: DGraph: A Large-Scale Financial Dataset for Graph Anomaly Detection
ABSTRACT: Graph Anomaly Detection (GAD) has recently become a hot research spot due to
its practicability and theoretical value. Since GAD emphasizes the application
and the rarity of anomalous samples, enriching the varieties of its datasets is
fundamental work. Thus, this paper present DGraph, a real-world dynamic graph
in the finance domain. DGraph overcomes many limitations of current GAD
datasets. It contains about 3M nodes, 4M dynamic edges, and 1M ground-truth
nodes. We provide a comprehensive observation of DGraph, revealing that
anomalous nodes and normal nodes generally have different structures, neighbor
distribution, and temporal dynamics. Moreover, it suggests that unlabeled nodes
are also essential for detecting fraudsters. Furthermore, we conduct extensive
experiments on DGraph. Observation and experiments demonstrate that DGraph is
propulsive to advance GAD research and enable in-depth exploration of anomalous
nodes.
|
2207.04396 | Minji Yoon | Minji Yoon, Yue Wu, John Palowitch, Bryan Perozzi, Ruslan
Salakhutdinov | Graph Generative Model for Benchmarking Graph Neural Networks | null | null | null | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | As the field of Graph Neural Networks (GNN) continues to grow, it experiences
a corresponding increase in the need for large, real-world datasets to train
and test new GNN models on challenging, realistic problems. Unfortunately, such
graph datasets are often generated from online, highly privacy-restricted
ecosystems, which makes research and development on these datasets hard, if not
impossible. This greatly reduces the amount of benchmark graphs available to
researchers, causing the field to rely only on a handful of publicly-available
datasets. To address this problem, we introduce a novel graph generative model,
Computation Graph Transformer (CGT) that learns and reproduces the distribution
of real-world graphs in a privacy-controlled way. More specifically, CGT (1)
generates effective benchmark graphs on which GNNs show similar task
performance as on the source graphs, (2) scales to process large-scale graphs,
(3) incorporates off-the-shelf privacy modules to guarantee end-user privacy of
the generated graph. Extensive experiments across a vast body of graph
generative models show that only our model can successfully generate
privacy-controlled, synthetic substitutes of large-scale real-world graphs that
can be effectively used to benchmark GNN models.
| [
{
"version": "v1",
"created": "Sun, 10 Jul 2022 06:42:02 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Oct 2022 23:37:49 GMT"
},
{
"version": "v3",
"created": "Mon, 20 Feb 2023 14:12:53 GMT"
},
{
"version": "v4",
"created": "Fri, 9 Jun 2023 11:52:42 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Yoon",
"Minji",
""
],
[
"Wu",
"Yue",
""
],
[
"Palowitch",
"John",
""
],
[
"Perozzi",
"Bryan",
""
],
[
"Salakhutdinov",
"Ruslan",
""
]
] | TITLE: Graph Generative Model for Benchmarking Graph Neural Networks
ABSTRACT: As the field of Graph Neural Networks (GNN) continues to grow, it experiences
a corresponding increase in the need for large, real-world datasets to train
and test new GNN models on challenging, realistic problems. Unfortunately, such
graph datasets are often generated from online, highly privacy-restricted
ecosystems, which makes research and development on these datasets hard, if not
impossible. This greatly reduces the amount of benchmark graphs available to
researchers, causing the field to rely only on a handful of publicly-available
datasets. To address this problem, we introduce a novel graph generative model,
Computation Graph Transformer (CGT) that learns and reproduces the distribution
of real-world graphs in a privacy-controlled way. More specifically, CGT (1)
generates effective benchmark graphs on which GNNs show similar task
performance as on the source graphs, (2) scales to process large-scale graphs,
(3) incorporates off-the-shelf privacy modules to guarantee end-user privacy of
the generated graph. Extensive experiments across a vast body of graph
generative models show that only our model can successfully generate
privacy-controlled, synthetic substitutes of large-scale real-world graphs that
can be effectively used to benchmark GNN models.
|
2207.12497 | Beepul Bharti | Beepul Bharti, Paul Yi, Jeremias Sulam | Estimating and Controlling for Equalized Odds via Sensitive Attribute
Predictors | null | null | null | null | cs.LG cs.CY | http://creativecommons.org/licenses/by/4.0/ | As the use of machine learning models in real world high-stakes decision
settings continues to grow, it is highly important that we are able to audit
and control for any potential fairness violations these models may exhibit
towards certain groups. To do so, one naturally requires access to sensitive
attributes, such as demographics, gender, or other potentially sensitive
features that determine group membership. Unfortunately, in many settings, this
information is often unavailable. In this work we study the well known
\emph{equalized odds} (EOD) definition of fairness. In a setting without
sensitive attributes, we first provide tight and computable upper bounds for
the EOD violation of a predictor. These bounds precisely reflect the worst
possible EOD violation. Second, we demonstrate how one can provably control the
worst-case EOD by a new post-processing correction method. Our results
characterize when directly controlling for EOD with respect to the predicted
sensitive attributes is -- and when is not -- optimal when it comes to
controlling worst-case EOD. Our results hold under assumptions that are milder
than previous works, and we illustrate these results with experiments on
synthetic and real datasets.
| [
{
"version": "v1",
"created": "Mon, 25 Jul 2022 19:56:02 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Oct 2022 20:48:37 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Feb 2023 19:11:17 GMT"
},
{
"version": "v4",
"created": "Fri, 9 Jun 2023 00:06:23 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Bharti",
"Beepul",
""
],
[
"Yi",
"Paul",
""
],
[
"Sulam",
"Jeremias",
""
]
] | TITLE: Estimating and Controlling for Equalized Odds via Sensitive Attribute
Predictors
ABSTRACT: As the use of machine learning models in real world high-stakes decision
settings continues to grow, it is highly important that we are able to audit
and control for any potential fairness violations these models may exhibit
towards certain groups. To do so, one naturally requires access to sensitive
attributes, such as demographics, gender, or other potentially sensitive
features that determine group membership. Unfortunately, in many settings, this
information is often unavailable. In this work we study the well known
\emph{equalized odds} (EOD) definition of fairness. In a setting without
sensitive attributes, we first provide tight and computable upper bounds for
the EOD violation of a predictor. These bounds precisely reflect the worst
possible EOD violation. Second, we demonstrate how one can provably control the
worst-case EOD by a new post-processing correction method. Our results
characterize when directly controlling for EOD with respect to the predicted
sensitive attributes is -- and when is not -- optimal when it comes to
controlling worst-case EOD. Our results hold under assumptions that are milder
than previous works, and we illustrate these results with experiments on
synthetic and real datasets.
|
2209.00439 | Shigehiko Schamoni | Shigehiko Schamoni, Michael Hagmann, Stefan Riezler | Ensembling Neural Networks for Improved Prediction and Privacy in Early
Diagnosis of Sepsis | Accepted at MLHC 2022 | Proceedings of the 7th Machine Learning for Healthcare Conference,
PMLR 182:123-145, 2022 | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Ensembling neural networks is a long-standing technique for improving the
generalization error of neural networks by combining networks with orthogonal
properties via a committee decision. We show that this technique is an ideal
fit for machine learning on medical data: First, ensembles are amenable to
parallel and asynchronous learning, thus enabling efficient training of
patient-specific component neural networks. Second, building on the idea of
minimizing generalization error by selecting uncorrelated patient-specific
networks, we show that one can build an ensemble of a few selected
patient-specific models that outperforms a single model trained on much larger
pooled datasets. Third, the non-iterative ensemble combination step is an
optimal low-dimensional entry point to apply output perturbation to guarantee
the privacy of the patient-specific networks. We exemplify our framework of
differentially private ensembles on the task of early prediction of sepsis,
using real-life intensive care unit data labeled by clinical experts.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2022 13:24:14 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Schamoni",
"Shigehiko",
""
],
[
"Hagmann",
"Michael",
""
],
[
"Riezler",
"Stefan",
""
]
] | TITLE: Ensembling Neural Networks for Improved Prediction and Privacy in Early
Diagnosis of Sepsis
ABSTRACT: Ensembling neural networks is a long-standing technique for improving the
generalization error of neural networks by combining networks with orthogonal
properties via a committee decision. We show that this technique is an ideal
fit for machine learning on medical data: First, ensembles are amenable to
parallel and asynchronous learning, thus enabling efficient training of
patient-specific component neural networks. Second, building on the idea of
minimizing generalization error by selecting uncorrelated patient-specific
networks, we show that one can build an ensemble of a few selected
patient-specific models that outperforms a single model trained on much larger
pooled datasets. Third, the non-iterative ensemble combination step is an
optimal low-dimensional entry point to apply output perturbation to guarantee
the privacy of the patient-specific networks. We exemplify our framework of
differentially private ensembles on the task of early prediction of sepsis,
using real-life intensive care unit data labeled by clinical experts.
|
2210.02871 | Seanie Lee | Seanie Lee, Minki Kang, Juho Lee, Sung Ju Hwang, Kenji Kawaguchi | Self-Distillation for Further Pre-training of Transformers | ICLR 2023 | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Pre-training a large transformer model on a massive amount of unlabeled data
and fine-tuning it on labeled datasets for diverse downstream tasks has proven
to be a successful strategy, for a variety of vision and natural language
processing tasks. However, direct fine-tuning of the pre-trained model may be
suboptimal if there exist large discrepancies across data domains for
pre-training and fine-tuning. To tackle this issue, several previous studies
have proposed further pre-training strategies, where we continue to pre-train
the model on the target unlabeled dataset before fine-tuning. However, all of
them solely focus on language models and we empirically find that a Vision
Transformer is vulnerable to overfitting as we continue to pretrain the model
on target unlabeled data. In order to tackle this limitation, we propose
self-distillation as a regularization for a further pre-training stage.
Specifically, we first further pre-train the initial pre-trained model on the
target unlabeled data and then consider it as a teacher for self-distillation.
Then we take the same initial pre-trained model as a student and enforce its
hidden representations to be close to those of the teacher while optimizing the
student with a masked auto-encoding objective. We empirically validate the
efficacy of self-distillation on a variety of benchmark datasets for image and
text classification tasks. Experimentally, we show that our proposed method
outperforms all the relevant baselines. Theoretically, we analyze the proposed
method with a simplified model to understand how self-distillation for further
pre-training can potentially help improve the performance of the downstream
tasks.
| [
{
"version": "v1",
"created": "Fri, 30 Sep 2022 02:25:12 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Mar 2023 07:07:10 GMT"
},
{
"version": "v3",
"created": "Fri, 9 Jun 2023 08:57:07 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lee",
"Seanie",
""
],
[
"Kang",
"Minki",
""
],
[
"Lee",
"Juho",
""
],
[
"Hwang",
"Sung Ju",
""
],
[
"Kawaguchi",
"Kenji",
""
]
] | TITLE: Self-Distillation for Further Pre-training of Transformers
ABSTRACT: Pre-training a large transformer model on a massive amount of unlabeled data
and fine-tuning it on labeled datasets for diverse downstream tasks has proven
to be a successful strategy, for a variety of vision and natural language
processing tasks. However, direct fine-tuning of the pre-trained model may be
suboptimal if there exist large discrepancies across data domains for
pre-training and fine-tuning. To tackle this issue, several previous studies
have proposed further pre-training strategies, where we continue to pre-train
the model on the target unlabeled dataset before fine-tuning. However, all of
them solely focus on language models and we empirically find that a Vision
Transformer is vulnerable to overfitting as we continue to pretrain the model
on target unlabeled data. In order to tackle this limitation, we propose
self-distillation as a regularization for a further pre-training stage.
Specifically, we first further pre-train the initial pre-trained model on the
target unlabeled data and then consider it as a teacher for self-distillation.
Then we take the same initial pre-trained model as a student and enforce its
hidden representations to be close to those of the teacher while optimizing the
student with a masked auto-encoding objective. We empirically validate the
efficacy of self-distillation on a variety of benchmark datasets for image and
text classification tasks. Experimentally, we show that our proposed method
outperforms all the relevant baselines. Theoretically, we analyze the proposed
method with a simplified model to understand how self-distillation for further
pre-training can potentially help improve the performance of the downstream
tasks.
|
2210.11780 | Tong Nie | Tong Nie, Guoyang Qin, Yunpeng Wang, Jian Sun | Correlating sparse sensing for large-scale traffic speed estimation: A
Laplacian-enhanced low-rank tensor kriging approach | null | null | 10.1016/j.trc.2023.104190 | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Traffic speed is central to characterizing the fluidity of the road network.
Many transportation applications rely on it, such as real-time navigation,
dynamic route planning, and congestion management. Rapid advances in sensing
and communication techniques make traffic speed detection easier than ever.
However, due to sparse deployment of static sensors or low penetration of
mobile sensors, speeds detected are incomplete and far from network-wide use.
In addition, sensors are prone to error or missing data due to various kinds of
reasons, speeds from these sensors can become highly noisy. These drawbacks
call for effective techniques to recover credible estimates from the incomplete
data. In this work, we first identify the issue as a spatiotemporal kriging
problem and propose a Laplacian enhanced low-rank tensor completion (LETC)
framework featuring both lowrankness and multi-dimensional correlations for
large-scale traffic speed kriging under limited observations. To be specific,
three types of speed correlation including temporal continuity, temporal
periodicity, and spatial proximity are carefully chosen and simultaneously
modeled by three different forms of graph Laplacian, named temporal graph
Fourier transform, generalized temporal consistency regularization, and
diffusion graph regularization. We then design an efficient solution algorithm
via several effective numeric techniques to scale up the proposed model to
network-wide kriging. By performing experiments on two public million-level
traffic speed datasets, we finally draw the conclusion and find our proposed
LETC achieves the state-of-the-art kriging performance even under low
observation rates, while at the same time saving more than half computing time
compared with baseline methods. Some insights into spatiotemporal traffic data
modeling and kriging at the network level are provided as well.
| [
{
"version": "v1",
"created": "Fri, 21 Oct 2022 07:25:57 GMT"
},
{
"version": "v2",
"created": "Wed, 1 Mar 2023 01:14:55 GMT"
},
{
"version": "v3",
"created": "Sun, 28 May 2023 06:56:56 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Nie",
"Tong",
""
],
[
"Qin",
"Guoyang",
""
],
[
"Wang",
"Yunpeng",
""
],
[
"Sun",
"Jian",
""
]
] | TITLE: Correlating sparse sensing for large-scale traffic speed estimation: A
Laplacian-enhanced low-rank tensor kriging approach
ABSTRACT: Traffic speed is central to characterizing the fluidity of the road network.
Many transportation applications rely on it, such as real-time navigation,
dynamic route planning, and congestion management. Rapid advances in sensing
and communication techniques make traffic speed detection easier than ever.
However, due to sparse deployment of static sensors or low penetration of
mobile sensors, speeds detected are incomplete and far from network-wide use.
In addition, sensors are prone to error or missing data due to various kinds of
reasons, speeds from these sensors can become highly noisy. These drawbacks
call for effective techniques to recover credible estimates from the incomplete
data. In this work, we first identify the issue as a spatiotemporal kriging
problem and propose a Laplacian enhanced low-rank tensor completion (LETC)
framework featuring both lowrankness and multi-dimensional correlations for
large-scale traffic speed kriging under limited observations. To be specific,
three types of speed correlation including temporal continuity, temporal
periodicity, and spatial proximity are carefully chosen and simultaneously
modeled by three different forms of graph Laplacian, named temporal graph
Fourier transform, generalized temporal consistency regularization, and
diffusion graph regularization. We then design an efficient solution algorithm
via several effective numeric techniques to scale up the proposed model to
network-wide kriging. By performing experiments on two public million-level
traffic speed datasets, we finally draw the conclusion and find our proposed
LETC achieves the state-of-the-art kriging performance even under low
observation rates, while at the same time saving more than half computing time
compared with baseline methods. Some insights into spatiotemporal traffic data
modeling and kriging at the network level are provided as well.
|
2210.12513 | Jiaming Chen | Jiaming Chen, Weixin Luo, Ran Song, Xiaolin Wei, Lin Ma, Wei Zhang | Learning Point-Language Hierarchical Alignment for 3D Visual Grounding | Champion on ECCV 2022 ScanRefer Challenge | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a novel hierarchical alignment model (HAM) that learns
multi-granularity visual and linguistic representations in an end-to-end
manner. We extract key points and proposal points to model 3D contexts and
instances, and propose point-language alignment with context modulation (PLACM)
mechanism, which learns to gradually align word-level and sentence-level
linguistic embeddings with visual representations, while the modulation with
the visual context captures latent informative relationships. To further
capture both global and local relationships, we propose a spatially
multi-granular modeling scheme that applies PLACM to both global and local
fields. Experimental results demonstrate the superiority of HAM, with
visualized results showing that it can dynamically model fine-grained visual
and linguistic representations. HAM outperforms existing methods by a
significant margin and achieves state-of-the-art performance on two publicly
available datasets, and won the championship in ECCV 2022 ScanRefer challenge.
Code is available at~\url{https://github.com/PPjmchen/HAM}.
| [
{
"version": "v1",
"created": "Sat, 22 Oct 2022 18:02:10 GMT"
},
{
"version": "v2",
"created": "Sun, 30 Oct 2022 09:22:05 GMT"
},
{
"version": "v3",
"created": "Mon, 5 Jun 2023 10:09:54 GMT"
},
{
"version": "v4",
"created": "Fri, 9 Jun 2023 04:06:39 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Chen",
"Jiaming",
""
],
[
"Luo",
"Weixin",
""
],
[
"Song",
"Ran",
""
],
[
"Wei",
"Xiaolin",
""
],
[
"Ma",
"Lin",
""
],
[
"Zhang",
"Wei",
""
]
] | TITLE: Learning Point-Language Hierarchical Alignment for 3D Visual Grounding
ABSTRACT: This paper presents a novel hierarchical alignment model (HAM) that learns
multi-granularity visual and linguistic representations in an end-to-end
manner. We extract key points and proposal points to model 3D contexts and
instances, and propose point-language alignment with context modulation (PLACM)
mechanism, which learns to gradually align word-level and sentence-level
linguistic embeddings with visual representations, while the modulation with
the visual context captures latent informative relationships. To further
capture both global and local relationships, we propose a spatially
multi-granular modeling scheme that applies PLACM to both global and local
fields. Experimental results demonstrate the superiority of HAM, with
visualized results showing that it can dynamically model fine-grained visual
and linguistic representations. HAM outperforms existing methods by a
significant margin and achieves state-of-the-art performance on two publicly
available datasets, and won the championship in ECCV 2022 ScanRefer challenge.
Code is available at~\url{https://github.com/PPjmchen/HAM}.
|
2210.14774 | Amine Bendahmane | Amine Bendahmane and Redouane Tlemsani | Unknown area exploration for robots with energy constraints using a
modified Butterfly Optimization Algorithm | null | Soft Computing volume 27, pages 3785-3804 (2023) | 10.1007/s00500-022-07530-w | null | cs.RO cs.AI cs.MA | http://creativecommons.org/licenses/by/4.0/ | Butterfly Optimization Algorithm (BOA) is a recent metaheuristic that has
been used in several optimization problems. In this paper, we propose a new
version of the algorithm (xBOA) based on the crossover operator and compare its
results to the original BOA and 3 other variants recently introduced in the
literature. We also proposed a framework for solving the unknown area
exploration problem with energy constraints using metaheuristics in both
single- and multi-robot scenarios. This framework allowed us to benchmark the
performances of different metaheuristics for the robotics exploration problem.
We conducted several experiments to validate this framework and used it to
compare the effectiveness of xBOA with wellknown metaheuristics used in the
literature through 5 evaluation criteria. Although BOA and xBOA are not optimal
in all these criteria, we found that BOA can be a good alternative to many
metaheuristics in terms of the exploration time, while xBOA is more robust to
local optima; has better fitness convergence; and achieves better exploration
rates than the original BOA and its other variants.
| [
{
"version": "v1",
"created": "Wed, 26 Oct 2022 15:15:19 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Bendahmane",
"Amine",
""
],
[
"Tlemsani",
"Redouane",
""
]
] | TITLE: Unknown area exploration for robots with energy constraints using a
modified Butterfly Optimization Algorithm
ABSTRACT: Butterfly Optimization Algorithm (BOA) is a recent metaheuristic that has
been used in several optimization problems. In this paper, we propose a new
version of the algorithm (xBOA) based on the crossover operator and compare its
results to the original BOA and 3 other variants recently introduced in the
literature. We also proposed a framework for solving the unknown area
exploration problem with energy constraints using metaheuristics in both
single- and multi-robot scenarios. This framework allowed us to benchmark the
performances of different metaheuristics for the robotics exploration problem.
We conducted several experiments to validate this framework and used it to
compare the effectiveness of xBOA with wellknown metaheuristics used in the
literature through 5 evaluation criteria. Although BOA and xBOA are not optimal
in all these criteria, we found that BOA can be a good alternative to many
metaheuristics in terms of the exploration time, while xBOA is more robust to
local optima; has better fitness convergence; and achieves better exploration
rates than the original BOA and its other variants.
|
2210.15398 | Shigehiko Schamoni | Tsz Kin Lam, Shigehiko Schamoni and Stefan Riezler | Make More of Your Data: Minimal Effort Data Augmentation for Automatic
Speech Recognition and Translation | Accepted at ICASSP 2023 | null | 10.1109/ICASSP49357.2023.10094564 | null | cs.CL cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Data augmentation is a technique to generate new training data based on
existing data. We evaluate the simple and cost-effective method of
concatenating the original data examples to build new training instances.
Continued training with such augmented data is able to improve off-the-shelf
Transformer and Conformer models that were optimized on the original data only.
We demonstrate considerable improvements on the LibriSpeech-960h test sets (WER
2.83 and 6.87 for test-clean and test-other), which carry over to models
combined with shallow fusion (WER 2.55 and 6.27). Our method of continued
training also leads to improvements of up to 0.9 WER on the ASR part of
CoVoST-2 for four non English languages, and we observe that the gains are
highly dependent on the size of the original training data. We compare
different concatenation strategies and found that our method does not need
speaker information to achieve its improvements. Finally, we demonstrate on two
datasets that our methods also works for speech translation tasks.
| [
{
"version": "v1",
"created": "Thu, 27 Oct 2022 13:01:01 GMT"
},
{
"version": "v2",
"created": "Fri, 14 Apr 2023 14:06:42 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lam",
"Tsz Kin",
""
],
[
"Schamoni",
"Shigehiko",
""
],
[
"Riezler",
"Stefan",
""
]
] | TITLE: Make More of Your Data: Minimal Effort Data Augmentation for Automatic
Speech Recognition and Translation
ABSTRACT: Data augmentation is a technique to generate new training data based on
existing data. We evaluate the simple and cost-effective method of
concatenating the original data examples to build new training instances.
Continued training with such augmented data is able to improve off-the-shelf
Transformer and Conformer models that were optimized on the original data only.
We demonstrate considerable improvements on the LibriSpeech-960h test sets (WER
2.83 and 6.87 for test-clean and test-other), which carry over to models
combined with shallow fusion (WER 2.55 and 6.27). Our method of continued
training also leads to improvements of up to 0.9 WER on the ASR part of
CoVoST-2 for four non English languages, and we observe that the gains are
highly dependent on the size of the original training data. We compare
different concatenation strategies and found that our method does not need
speaker information to achieve its improvements. Finally, we demonstrate on two
datasets that our methods also works for speech translation tasks.
|
2211.06530 | Christopher A. Choquette-Choo | Christopher A. Choquette-Choo, H. Brendan McMahan, Keith Rush, and
Abhradeep Thakurta | Multi-Epoch Matrix Factorization Mechanisms for Private Machine Learning | 9 pages main-text, 3 figures. 40 pages with 13 figures total | null | null | null | cs.LG cs.CR cs.DS stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce new differentially private (DP) mechanisms for gradient-based
machine learning (ML) with multiple passes (epochs) over a dataset,
substantially improving the achievable privacy-utility-computation tradeoffs.
We formalize the problem of DP mechanisms for adaptive streams with multiple
participations and introduce a non-trivial extension of online matrix
factorization DP mechanisms to our setting. This includes establishing the
necessary theory for sensitivity calculations and efficient computation of
optimal matrices. For some applications like $>\!\! 10,000$ SGD steps, applying
these optimal techniques becomes computationally expensive. We thus design an
efficient Fourier-transform-based mechanism with only a minor utility loss.
Extensive empirical evaluation on both example-level DP for image
classification and user-level DP for language modeling demonstrate substantial
improvements over all previous methods, including the widely-used DP-SGD .
Though our primary application is to ML, our main DP results are applicable to
arbitrary linear queries and hence may have much broader applicability.
| [
{
"version": "v1",
"created": "Sat, 12 Nov 2022 00:41:11 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Jun 2023 19:59:03 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Choquette-Choo",
"Christopher A.",
""
],
[
"McMahan",
"H. Brendan",
""
],
[
"Rush",
"Keith",
""
],
[
"Thakurta",
"Abhradeep",
""
]
] | TITLE: Multi-Epoch Matrix Factorization Mechanisms for Private Machine Learning
ABSTRACT: We introduce new differentially private (DP) mechanisms for gradient-based
machine learning (ML) with multiple passes (epochs) over a dataset,
substantially improving the achievable privacy-utility-computation tradeoffs.
We formalize the problem of DP mechanisms for adaptive streams with multiple
participations and introduce a non-trivial extension of online matrix
factorization DP mechanisms to our setting. This includes establishing the
necessary theory for sensitivity calculations and efficient computation of
optimal matrices. For some applications like $>\!\! 10,000$ SGD steps, applying
these optimal techniques becomes computationally expensive. We thus design an
efficient Fourier-transform-based mechanism with only a minor utility loss.
Extensive empirical evaluation on both example-level DP for image
classification and user-level DP for language modeling demonstrate substantial
improvements over all previous methods, including the widely-used DP-SGD .
Though our primary application is to ML, our main DP results are applicable to
arbitrary linear queries and hence may have much broader applicability.
|
2211.12252 | Hicham Johra Dr. | H. Li, H. Johra, F. de Andrade Pereira, T. Hong, J. Le Dreau, A.
Maturo, M. Wei, Y. Liu, A. Saberi-Derakhtenjani, Z. Nagy, A.
Marszal-Pomianowska, D. Finn, S. Miyata, K. Kaspar, K. Nweye, Z. O Neill, F.
Pallonetto, B. Dong | Data-Driven Key Performance Indicators and Datasets for Building Energy
Flexibility: A Review and Perspectives | 30 pages, 14 figures, 4 tables | Applied Energy 2023, Volume 343 | 10.1016/j.apenergy.2023.121217 | 121217 | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | Energy flexibility, through short-term demand-side management (DSM) and
energy storage technologies, is now seen as a major key to balancing the
fluctuating supply in different energy grids with the energy demand of
buildings. This is especially important when considering the intermittent
nature of ever-growing renewable energy production, as well as the increasing
dynamics of electricity demand in buildings. This paper provides a holistic
review of (1) data-driven energy flexibility key performance indicators (KPIs)
for buildings in the operational phase and (2) open datasets that can be used
for testing energy flexibility KPIs. The review identifies a total of 81
data-driven KPIs from 91 recent publications. These KPIs were categorized and
analyzed according to their type, complexity, scope, key stakeholders, data
requirement, baseline requirement, resolution, and popularity. Moreover, 330
building datasets were collected and evaluated. Of those, 16 were deemed
adequate to feature building performing demand response or building-to-grid
(B2G) services. The DSM strategy, building scope, grid type, control strategy,
needed data features, and usability of these selected 16 datasets were
analyzed. This review reveals future opportunities to address limitations in
the existing literature: (1) developing new data-driven methodologies to
specifically evaluate different energy flexibility strategies and B2G services
of existing buildings; (2) developing baseline-free KPIs that could be
calculated from easily accessible building sensors and meter data; (3) devoting
non-engineering efforts to promote building energy flexibility, such as
designing utility programs, standardizing energy flexibility quantification and
verification processes; and (4) curating datasets with proper description for
energy flexibility assessments.
| [
{
"version": "v1",
"created": "Tue, 22 Nov 2022 13:12:29 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 14:51:36 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Li",
"H.",
""
],
[
"Johra",
"H.",
""
],
[
"Pereira",
"F. de Andrade",
""
],
[
"Hong",
"T.",
""
],
[
"Dreau",
"J. Le",
""
],
[
"Maturo",
"A.",
""
],
[
"Wei",
"M.",
""
],
[
"Liu",
"Y.",
""
],
[
"Saberi-Derakhtenjani",
"A.",
""
],
[
"Nagy",
"Z.",
""
],
[
"Marszal-Pomianowska",
"A.",
""
],
[
"Finn",
"D.",
""
],
[
"Miyata",
"S.",
""
],
[
"Kaspar",
"K.",
""
],
[
"Nweye",
"K.",
""
],
[
"Neill",
"Z. O",
""
],
[
"Pallonetto",
"F.",
""
],
[
"Dong",
"B.",
""
]
] | TITLE: Data-Driven Key Performance Indicators and Datasets for Building Energy
Flexibility: A Review and Perspectives
ABSTRACT: Energy flexibility, through short-term demand-side management (DSM) and
energy storage technologies, is now seen as a major key to balancing the
fluctuating supply in different energy grids with the energy demand of
buildings. This is especially important when considering the intermittent
nature of ever-growing renewable energy production, as well as the increasing
dynamics of electricity demand in buildings. This paper provides a holistic
review of (1) data-driven energy flexibility key performance indicators (KPIs)
for buildings in the operational phase and (2) open datasets that can be used
for testing energy flexibility KPIs. The review identifies a total of 81
data-driven KPIs from 91 recent publications. These KPIs were categorized and
analyzed according to their type, complexity, scope, key stakeholders, data
requirement, baseline requirement, resolution, and popularity. Moreover, 330
building datasets were collected and evaluated. Of those, 16 were deemed
adequate to feature building performing demand response or building-to-grid
(B2G) services. The DSM strategy, building scope, grid type, control strategy,
needed data features, and usability of these selected 16 datasets were
analyzed. This review reveals future opportunities to address limitations in
the existing literature: (1) developing new data-driven methodologies to
specifically evaluate different energy flexibility strategies and B2G services
of existing buildings; (2) developing baseline-free KPIs that could be
calculated from easily accessible building sensors and meter data; (3) devoting
non-engineering efforts to promote building energy flexibility, such as
designing utility programs, standardizing energy flexibility quantification and
verification processes; and (4) curating datasets with proper description for
energy flexibility assessments.
|
2211.15355 | Wenxuan Zhu | Wenxuan Zhu, Chao Yu, Qiang Zhang | Causal Deep Reinforcement Learning Using Observational Data | null | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep reinforcement learning (DRL) requires the collection of interventional
data, which is sometimes expensive and even unethical in the real world, such
as in the autonomous driving and the medical field. Offline reinforcement
learning promises to alleviate this issue by exploiting the vast amount of
observational data available in the real world. However, observational data may
mislead the learning agent to undesirable outcomes if the behavior policy that
generates the data depends on unobserved random variables (i.e., confounders).
In this paper, we propose two deconfounding methods in DRL to address this
problem. The methods first calculate the importance degree of different samples
based on the causal inference technique, and then adjust the impact of
different samples on the loss function by reweighting or resampling the offline
dataset to ensure its unbiasedness. These deconfounding methods can be flexibly
combined with existing model-free DRL algorithms such as soft actor-critic and
deep Q-learning, provided that a weak condition can be satisfied by the loss
functions of these algorithms. We prove the effectiveness of our deconfounding
methods and validate them experimentally.
| [
{
"version": "v1",
"created": "Mon, 28 Nov 2022 14:34:39 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 17:03:15 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Zhu",
"Wenxuan",
""
],
[
"Yu",
"Chao",
""
],
[
"Zhang",
"Qiang",
""
]
] | TITLE: Causal Deep Reinforcement Learning Using Observational Data
ABSTRACT: Deep reinforcement learning (DRL) requires the collection of interventional
data, which is sometimes expensive and even unethical in the real world, such
as in the autonomous driving and the medical field. Offline reinforcement
learning promises to alleviate this issue by exploiting the vast amount of
observational data available in the real world. However, observational data may
mislead the learning agent to undesirable outcomes if the behavior policy that
generates the data depends on unobserved random variables (i.e., confounders).
In this paper, we propose two deconfounding methods in DRL to address this
problem. The methods first calculate the importance degree of different samples
based on the causal inference technique, and then adjust the impact of
different samples on the loss function by reweighting or resampling the offline
dataset to ensure its unbiasedness. These deconfounding methods can be flexibly
combined with existing model-free DRL algorithms such as soft actor-critic and
deep Q-learning, provided that a weak condition can be satisfied by the loss
functions of these algorithms. We prove the effectiveness of our deconfounding
methods and validate them experimentally.
|
2212.02191 | Bo Li | Bo Li, Mikkel N. Schmidt, Tommy S. Alstr{\o}m, Sebastian U. Stich | On the effectiveness of partial variance reduction in federated learning
with heterogeneous data | Accepted to CVPR 2023 | null | null | null | cs.LG cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data heterogeneity across clients is a key challenge in federated learning.
Prior works address this by either aligning client and server models or using
control variates to correct client model drift. Although these methods achieve
fast convergence in convex or simple non-convex problems, the performance in
over-parameterized models such as deep neural networks is lacking. In this
paper, we first revisit the widely used FedAvg algorithm in a deep neural
network to understand how data heterogeneity influences the gradient updates
across the neural network layers. We observe that while the feature extraction
layers are learned efficiently by FedAvg, the substantial diversity of the
final classification layers across clients impedes the performance. Motivated
by this, we propose to correct model drift by variance reduction only on the
final layers. We demonstrate that this significantly outperforms existing
benchmarks at a similar or lower communication cost. We furthermore provide
proof for the convergence rate of our algorithm.
| [
{
"version": "v1",
"created": "Mon, 5 Dec 2022 11:56:35 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 11:13:54 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Li",
"Bo",
""
],
[
"Schmidt",
"Mikkel N.",
""
],
[
"Alstrøm",
"Tommy S.",
""
],
[
"Stich",
"Sebastian U.",
""
]
] | TITLE: On the effectiveness of partial variance reduction in federated learning
with heterogeneous data
ABSTRACT: Data heterogeneity across clients is a key challenge in federated learning.
Prior works address this by either aligning client and server models or using
control variates to correct client model drift. Although these methods achieve
fast convergence in convex or simple non-convex problems, the performance in
over-parameterized models such as deep neural networks is lacking. In this
paper, we first revisit the widely used FedAvg algorithm in a deep neural
network to understand how data heterogeneity influences the gradient updates
across the neural network layers. We observe that while the feature extraction
layers are learned efficiently by FedAvg, the substantial diversity of the
final classification layers across clients impedes the performance. Motivated
by this, we propose to correct model drift by variance reduction only on the
final layers. We demonstrate that this significantly outperforms existing
benchmarks at a similar or lower communication cost. We furthermore provide
proof for the convergence rate of our algorithm.
|
2301.02297 | Thomas Hitchcox | Thomas Hitchcox and James Richard Forbes | Improving Self-Consistency in Underwater Mapping Through Laser-Based
Loop Closure (Extended) | 26 pages, 18 figures. V2 correct Table III x2 parameter values, Table
VIII 'INS' values, and equation A.28 | IEEE Transactions on Robotics, vol. 39, no. 3, pp. 1873-1892, 2023 | 10.1109/TRO.2022.3229842 | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate, self-consistent bathymetric maps are needed to monitor changes in
subsea environments and infrastructure. These maps are increasingly collected
by underwater vehicles, and mapping requires an accurate vehicle navigation
solution. Commercial off-the-shelf (COTS) navigation solutions for underwater
vehicles often rely on external acoustic sensors for localization, however
survey-grade acoustic sensors are expensive to deploy and limit the range of
the vehicle. Techniques from the field of simultaneous localization and
mapping, particularly loop closures, can improve the quality of the navigation
solution over dead-reckoning, but are difficult to integrate into COTS
navigation systems. This work presents a method to improve the self-consistency
of bathymetric maps by smoothly integrating loop-closure measurements into the
state estimate produced by a commercial subsea navigation system. Integration
is done using a white-noise-on-acceleration motion prior, without access to raw
sensor measurements or proprietary models. Improvements in map self-consistency
are shown for both simulated and experimental datasets, including a 3D scan of
an underwater shipwreck in Wiarton, Ontario, Canada.
| [
{
"version": "v1",
"created": "Thu, 5 Jan 2023 21:14:28 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 15:49:03 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Hitchcox",
"Thomas",
""
],
[
"Forbes",
"James Richard",
""
]
] | TITLE: Improving Self-Consistency in Underwater Mapping Through Laser-Based
Loop Closure (Extended)
ABSTRACT: Accurate, self-consistent bathymetric maps are needed to monitor changes in
subsea environments and infrastructure. These maps are increasingly collected
by underwater vehicles, and mapping requires an accurate vehicle navigation
solution. Commercial off-the-shelf (COTS) navigation solutions for underwater
vehicles often rely on external acoustic sensors for localization, however
survey-grade acoustic sensors are expensive to deploy and limit the range of
the vehicle. Techniques from the field of simultaneous localization and
mapping, particularly loop closures, can improve the quality of the navigation
solution over dead-reckoning, but are difficult to integrate into COTS
navigation systems. This work presents a method to improve the self-consistency
of bathymetric maps by smoothly integrating loop-closure measurements into the
state estimate produced by a commercial subsea navigation system. Integration
is done using a white-noise-on-acceleration motion prior, without access to raw
sensor measurements or proprietary models. Improvements in map self-consistency
are shown for both simulated and experimental datasets, including a 3D scan of
an underwater shipwreck in Wiarton, Ontario, Canada.
|
2302.00163 | Amin Farajzadeh | Amin Farajzadeh, Animesh Yadav, Omid Abbasi, Wael Jaafar, Halim
Yanikomeroglu | FLSTRA: Federated Learning in Stratosphere | Accepted to IEEE Transactions on Wireless Communications | null | null | null | cs.NI cs.LG eess.SP math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a federated learning (FL) in stratosphere (FLSTRA) system, where a
high altitude platform station (HAPS) facilitates a large number of terrestrial
clients to collaboratively learn a global model without sharing the training
data. FLSTRA overcomes the challenges faced by FL in terrestrial networks, such
as slow convergence and high communication delay due to limited client
participation and multi-hop communications. HAPS leverages its altitude and
size to allow the participation of more clients with line-of-sight (LOS) links
and the placement of a powerful server. However, handling many clients at once
introduces computing and transmission delays. Thus, we aim to obtain a
delay-accuracy trade-off for FLSTRA. Specifically, we first develop a joint
client selection and resource allocation algorithm for uplink and downlink to
minimize the FL delay subject to the energy and quality-of-service (QoS)
constraints. Second, we propose a communication and computation resource-aware
(CCRA-FL) algorithm to achieve the target FL accuracy while deriving an upper
bound for its convergence rate. The formulated problem is non-convex; thus, we
propose an iterative algorithm to solve it. Simulation results demonstrate the
effectiveness of the proposed FLSTRA system, compared to terrestrial
benchmarks, in terms of FL delay and accuracy.
| [
{
"version": "v1",
"created": "Wed, 1 Feb 2023 00:52:55 GMT"
},
{
"version": "v2",
"created": "Tue, 9 May 2023 18:35:34 GMT"
},
{
"version": "v3",
"created": "Fri, 9 Jun 2023 14:26:29 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Farajzadeh",
"Amin",
""
],
[
"Yadav",
"Animesh",
""
],
[
"Abbasi",
"Omid",
""
],
[
"Jaafar",
"Wael",
""
],
[
"Yanikomeroglu",
"Halim",
""
]
] | TITLE: FLSTRA: Federated Learning in Stratosphere
ABSTRACT: We propose a federated learning (FL) in stratosphere (FLSTRA) system, where a
high altitude platform station (HAPS) facilitates a large number of terrestrial
clients to collaboratively learn a global model without sharing the training
data. FLSTRA overcomes the challenges faced by FL in terrestrial networks, such
as slow convergence and high communication delay due to limited client
participation and multi-hop communications. HAPS leverages its altitude and
size to allow the participation of more clients with line-of-sight (LOS) links
and the placement of a powerful server. However, handling many clients at once
introduces computing and transmission delays. Thus, we aim to obtain a
delay-accuracy trade-off for FLSTRA. Specifically, we first develop a joint
client selection and resource allocation algorithm for uplink and downlink to
minimize the FL delay subject to the energy and quality-of-service (QoS)
constraints. Second, we propose a communication and computation resource-aware
(CCRA-FL) algorithm to achieve the target FL accuracy while deriving an upper
bound for its convergence rate. The formulated problem is non-convex; thus, we
propose an iterative algorithm to solve it. Simulation results demonstrate the
effectiveness of the proposed FLSTRA system, compared to terrestrial
benchmarks, in terms of FL delay and accuracy.
|
2302.03806 | Vasilis Kontonis | Vasilis Kontonis, Fotis Iliopoulos, Khoa Trinh, Cenk Baykal, Gaurav
Menghani, Erik Vee | SLaM: Student-Label Mixing for Distillation with Unlabeled Examples | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Knowledge distillation with unlabeled examples is a powerful training
paradigm for generating compact and lightweight student models in applications
where the amount of labeled data is limited but one has access to a large pool
of unlabeled data. In this setting, a large teacher model generates ``soft''
pseudo-labels for the unlabeled dataset which are then used for training the
student model. Despite its success in a wide variety of applications, a
shortcoming of this approach is that the teacher's pseudo-labels are often
noisy, leading to impaired student performance. In this paper, we present a
principled method for knowledge distillation with unlabeled examples that we
call Student-Label Mixing (SLaM) and we show that it consistently improves over
prior approaches by evaluating it on several standard benchmarks. Finally, we
show that SLaM comes with theoretical guarantees; along the way we give an
algorithm improving the best-known sample complexity for learning halfspaces
with margin under random classification noise, and provide the first
convergence analysis for so-called ``forward loss-adjustment" methods.
| [
{
"version": "v1",
"created": "Wed, 8 Feb 2023 00:14:44 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Jun 2023 18:00:24 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Kontonis",
"Vasilis",
""
],
[
"Iliopoulos",
"Fotis",
""
],
[
"Trinh",
"Khoa",
""
],
[
"Baykal",
"Cenk",
""
],
[
"Menghani",
"Gaurav",
""
],
[
"Vee",
"Erik",
""
]
] | TITLE: SLaM: Student-Label Mixing for Distillation with Unlabeled Examples
ABSTRACT: Knowledge distillation with unlabeled examples is a powerful training
paradigm for generating compact and lightweight student models in applications
where the amount of labeled data is limited but one has access to a large pool
of unlabeled data. In this setting, a large teacher model generates ``soft''
pseudo-labels for the unlabeled dataset which are then used for training the
student model. Despite its success in a wide variety of applications, a
shortcoming of this approach is that the teacher's pseudo-labels are often
noisy, leading to impaired student performance. In this paper, we present a
principled method for knowledge distillation with unlabeled examples that we
call Student-Label Mixing (SLaM) and we show that it consistently improves over
prior approaches by evaluating it on several standard benchmarks. Finally, we
show that SLaM comes with theoretical guarantees; along the way we give an
algorithm improving the best-known sample complexity for learning halfspaces
with margin under random classification noise, and provide the first
convergence analysis for so-called ``forward loss-adjustment" methods.
|
2302.10697 | Binwei Xu | Binwei Xu, Haoran Liang, Weihua Gong, Ronghua Liang, Peng Chen | A Visual Representation-guided Framework with Global Affinity for Weakly
Supervised Salient Object Detection | null | null | 10.1109/TCSVT.2023.3284076 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fully supervised salient object detection (SOD) methods have made
considerable progress in performance, yet these models rely heavily on
expensive pixel-wise labels. Recently, to achieve a trade-off between labeling
burden and performance, scribble-based SOD methods have attracted increasing
attention. Previous scribble-based models directly implement the SOD task only
based on SOD training data with limited information, it is extremely difficult
for them to understand the image and further achieve a superior SOD task. In
this paper, we propose a simple yet effective framework guided by general
visual representations with rich contextual semantic knowledge for
scribble-based SOD. These general visual representations are generated by
self-supervised learning based on large-scale unlabeled datasets. Our framework
consists of a task-related encoder, a general visual module, and an information
integration module to efficiently combine the general visual representations
with task-related features to perform the SOD task based on understanding the
contextual connections of images. Meanwhile, we propose a novel global semantic
affinity loss to guide the model to perceive the global structure of the
salient objects. Experimental results on five public benchmark datasets
demonstrate that our method, which only utilizes scribble annotations without
introducing any extra label, outperforms the state-of-the-art weakly supervised
SOD methods. Specifically, it outperforms the previous best scribble-based
method on all datasets with an average gain of 5.5% for max f-measure, 5.8% for
mean f-measure, 24% for MAE, and 3.1% for E-measure. Moreover, our method
achieves comparable or even superior performance to the state-of-the-art fully
supervised models.
| [
{
"version": "v1",
"created": "Tue, 21 Feb 2023 14:31:57 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 01:30:00 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Xu",
"Binwei",
""
],
[
"Liang",
"Haoran",
""
],
[
"Gong",
"Weihua",
""
],
[
"Liang",
"Ronghua",
""
],
[
"Chen",
"Peng",
""
]
] | TITLE: A Visual Representation-guided Framework with Global Affinity for Weakly
Supervised Salient Object Detection
ABSTRACT: Fully supervised salient object detection (SOD) methods have made
considerable progress in performance, yet these models rely heavily on
expensive pixel-wise labels. Recently, to achieve a trade-off between labeling
burden and performance, scribble-based SOD methods have attracted increasing
attention. Previous scribble-based models directly implement the SOD task only
based on SOD training data with limited information, it is extremely difficult
for them to understand the image and further achieve a superior SOD task. In
this paper, we propose a simple yet effective framework guided by general
visual representations with rich contextual semantic knowledge for
scribble-based SOD. These general visual representations are generated by
self-supervised learning based on large-scale unlabeled datasets. Our framework
consists of a task-related encoder, a general visual module, and an information
integration module to efficiently combine the general visual representations
with task-related features to perform the SOD task based on understanding the
contextual connections of images. Meanwhile, we propose a novel global semantic
affinity loss to guide the model to perceive the global structure of the
salient objects. Experimental results on five public benchmark datasets
demonstrate that our method, which only utilizes scribble annotations without
introducing any extra label, outperforms the state-of-the-art weakly supervised
SOD methods. Specifically, it outperforms the previous best scribble-based
method on all datasets with an average gain of 5.5% for max f-measure, 5.8% for
mean f-measure, 24% for MAE, and 3.1% for E-measure. Moreover, our method
achieves comparable or even superior performance to the state-of-the-art fully
supervised models.
|
2304.04740 | Aaron Lou | Aaron Lou and Stefano Ermon | Reflected Diffusion Models | ICML 2023 Camera Ready. Code available at
https://github.com/louaaron/Reflected-Diffusion | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | Score-based diffusion models learn to reverse a stochastic differential
equation that maps data to noise. However, for complex tasks, numerical error
can compound and result in highly unnatural samples. Previous work mitigates
this drift with thresholding, which projects to the natural data domain (such
as pixel space for images) after each diffusion step, but this leads to a
mismatch between the training and generative processes. To incorporate data
constraints in a principled manner, we present Reflected Diffusion Models,
which instead reverse a reflected stochastic differential equation evolving on
the support of the data. Our approach learns the perturbed score function
through a generalized score matching loss and extends key components of
standard diffusion models including diffusion guidance, likelihood-based
training, and ODE sampling. We also bridge the theoretical gap with
thresholding: such schemes are just discretizations of reflected SDEs. On
standard image benchmarks, our method is competitive with or surpasses the
state of the art without architectural modifications and, for classifier-free
guidance, our approach enables fast exact sampling with ODEs and produces more
faithful samples under high guidance weight.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 17:54:38 GMT"
},
{
"version": "v2",
"created": "Thu, 13 Apr 2023 17:01:37 GMT"
},
{
"version": "v3",
"created": "Thu, 8 Jun 2023 21:56:08 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lou",
"Aaron",
""
],
[
"Ermon",
"Stefano",
""
]
] | TITLE: Reflected Diffusion Models
ABSTRACT: Score-based diffusion models learn to reverse a stochastic differential
equation that maps data to noise. However, for complex tasks, numerical error
can compound and result in highly unnatural samples. Previous work mitigates
this drift with thresholding, which projects to the natural data domain (such
as pixel space for images) after each diffusion step, but this leads to a
mismatch between the training and generative processes. To incorporate data
constraints in a principled manner, we present Reflected Diffusion Models,
which instead reverse a reflected stochastic differential equation evolving on
the support of the data. Our approach learns the perturbed score function
through a generalized score matching loss and extends key components of
standard diffusion models including diffusion guidance, likelihood-based
training, and ODE sampling. We also bridge the theoretical gap with
thresholding: such schemes are just discretizations of reflected SDEs. On
standard image benchmarks, our method is competitive with or surpasses the
state of the art without architectural modifications and, for classifier-free
guidance, our approach enables fast exact sampling with ODEs and produces more
faithful samples under high guidance weight.
|
2304.13835 | Jack Urbanek | Jimmy Wei, Kurt Shuster, Arthur Szlam, Jason Weston, Jack Urbanek,
Mojtaba Komeili | Multi-Party Chat: Conversational Agents in Group Settings with Humans
and Models | null | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current dialogue research primarily studies pairwise (two-party)
conversations, and does not address the everyday setting where more than two
speakers converse together. In this work, we both collect and evaluate
multi-party conversations to study this more general case. We use the LIGHT
environment to construct grounded conversations, where each participant has an
assigned character to role-play. We thus evaluate the ability of language
models to act as one or more characters in such conversations. Models require
two skills that pairwise-trained models appear to lack: (1) being able to
decide when to talk; (2) producing coherent utterances grounded on multiple
characters. We compare models trained on our new dataset to existing
pairwise-trained dialogue models, as well as large language models with
few-shot prompting. We find that our new dataset, MultiLIGHT, which we will
publicly release, can help bring significant improvements in the group setting.
| [
{
"version": "v1",
"created": "Wed, 26 Apr 2023 21:41:17 GMT"
},
{
"version": "v2",
"created": "Wed, 7 Jun 2023 14:53:55 GMT"
},
{
"version": "v3",
"created": "Thu, 8 Jun 2023 21:45:46 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wei",
"Jimmy",
""
],
[
"Shuster",
"Kurt",
""
],
[
"Szlam",
"Arthur",
""
],
[
"Weston",
"Jason",
""
],
[
"Urbanek",
"Jack",
""
],
[
"Komeili",
"Mojtaba",
""
]
] | TITLE: Multi-Party Chat: Conversational Agents in Group Settings with Humans
and Models
ABSTRACT: Current dialogue research primarily studies pairwise (two-party)
conversations, and does not address the everyday setting where more than two
speakers converse together. In this work, we both collect and evaluate
multi-party conversations to study this more general case. We use the LIGHT
environment to construct grounded conversations, where each participant has an
assigned character to role-play. We thus evaluate the ability of language
models to act as one or more characters in such conversations. Models require
two skills that pairwise-trained models appear to lack: (1) being able to
decide when to talk; (2) producing coherent utterances grounded on multiple
characters. We compare models trained on our new dataset to existing
pairwise-trained dialogue models, as well as large language models with
few-shot prompting. We find that our new dataset, MultiLIGHT, which we will
publicly release, can help bring significant improvements in the group setting.
|
2305.01555 | Ningyu Zhang | Xin Xu, Yuqi Zhu, Xiaohan Wang, Ningyu Zhang | How to Unleash the Power of Large Language Models for Few-shot Relation
Extraction? | SustaiNLP Workshop@ACL 2023 | null | null | null | cs.CL cs.AI cs.DB cs.IR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Scaling language models have revolutionized widespread NLP tasks, yet little
comprehensively explored few-shot relation extraction with large language
models. In this paper, we investigate principal methodologies, in-context
learning and data generation, for few-shot relation extraction via GPT-3.5
through exhaustive experiments. To enhance few-shot performance, we further
propose task-related instructions and schema-constrained data generation. We
observe that in-context learning can achieve performance on par with previous
prompt learning approaches, and data generation with the large language model
can boost previous solutions to obtain new state-of-the-art few-shot results on
four widely-studied relation extraction datasets. We hope our work can inspire
future research for the capabilities of large language models in few-shot
relation extraction. Code is available in
https://github.com/zjunlp/DeepKE/tree/main/example/llm.
| [
{
"version": "v1",
"created": "Tue, 2 May 2023 15:55:41 GMT"
},
{
"version": "v2",
"created": "Wed, 3 May 2023 06:45:59 GMT"
},
{
"version": "v3",
"created": "Thu, 1 Jun 2023 06:51:58 GMT"
},
{
"version": "v4",
"created": "Fri, 9 Jun 2023 15:59:18 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Xu",
"Xin",
""
],
[
"Zhu",
"Yuqi",
""
],
[
"Wang",
"Xiaohan",
""
],
[
"Zhang",
"Ningyu",
""
]
] | TITLE: How to Unleash the Power of Large Language Models for Few-shot Relation
Extraction?
ABSTRACT: Scaling language models have revolutionized widespread NLP tasks, yet little
comprehensively explored few-shot relation extraction with large language
models. In this paper, we investigate principal methodologies, in-context
learning and data generation, for few-shot relation extraction via GPT-3.5
through exhaustive experiments. To enhance few-shot performance, we further
propose task-related instructions and schema-constrained data generation. We
observe that in-context learning can achieve performance on par with previous
prompt learning approaches, and data generation with the large language model
can boost previous solutions to obtain new state-of-the-art few-shot results on
four widely-studied relation extraction datasets. We hope our work can inspire
future research for the capabilities of large language models in few-shot
relation extraction. Code is available in
https://github.com/zjunlp/DeepKE/tree/main/example/llm.
|
2305.01904 | KiYoon Yoo | KiYoon Yoo, Wonhyuk Ahn, Jiho Jang, Nojun Kwak | Robust Multi-bit Natural Language Watermarking through Invariant
Features | ACL 2023 long | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent years have witnessed a proliferation of valuable original natural
language contents found in subscription-based media outlets, web novel
platforms, and outputs of large language models. However, these contents are
susceptible to illegal piracy and potential misuse without proper security
measures. This calls for a secure watermarking system to guarantee copyright
protection through leakage tracing or ownership identification. To effectively
combat piracy and protect copyrights, a multi-bit watermarking framework should
be able to embed adequate bits of information and extract the watermarks in a
robust manner despite possible corruption. In this work, we explore ways to
advance both payload and robustness by following a well-known proposition from
image watermarking and identify features in natural language that are invariant
to minor corruption. Through a systematic analysis of the possible sources of
errors, we further propose a corruption-resistant infill model. Our full method
improves upon the previous work on robustness by +16.8% point on average on
four datasets, three corruption types, and two corruption ratios. Code
available at https://github.com/bangawayoo/nlp-watermarking.
| [
{
"version": "v1",
"created": "Wed, 3 May 2023 05:37:30 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 07:17:14 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Yoo",
"KiYoon",
""
],
[
"Ahn",
"Wonhyuk",
""
],
[
"Jang",
"Jiho",
""
],
[
"Kwak",
"Nojun",
""
]
] | TITLE: Robust Multi-bit Natural Language Watermarking through Invariant
Features
ABSTRACT: Recent years have witnessed a proliferation of valuable original natural
language contents found in subscription-based media outlets, web novel
platforms, and outputs of large language models. However, these contents are
susceptible to illegal piracy and potential misuse without proper security
measures. This calls for a secure watermarking system to guarantee copyright
protection through leakage tracing or ownership identification. To effectively
combat piracy and protect copyrights, a multi-bit watermarking framework should
be able to embed adequate bits of information and extract the watermarks in a
robust manner despite possible corruption. In this work, we explore ways to
advance both payload and robustness by following a well-known proposition from
image watermarking and identify features in natural language that are invariant
to minor corruption. Through a systematic analysis of the possible sources of
errors, we further propose a corruption-resistant infill model. Our full method
improves upon the previous work on robustness by +16.8% point on average on
four datasets, three corruption types, and two corruption ratios. Code
available at https://github.com/bangawayoo/nlp-watermarking.
|
2305.03827 | Yufei Li | Yufei Li, Xiao Yu, Yanchi Liu, Haifeng Chen, Cong Liu | Uncertainty-Aware Bootstrap Learning for Joint Extraction on
Distantly-Supervised Data | ACL 2023 main conference short paper | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Jointly extracting entity pairs and their relations is challenging when
working on distantly-supervised data with ambiguous or noisy labels. To
mitigate such impact, we propose uncertainty-aware bootstrap learning, which is
motivated by the intuition that the higher uncertainty of an instance, the more
likely the model confidence is inconsistent with the ground truths.
Specifically, we first explore instance-level data uncertainty to create an
initial high-confident examples. Such subset serves as filtering noisy
instances and facilitating the model to converge fast at the early stage.
During bootstrap learning, we propose self-ensembling as a regularizer to
alleviate inter-model uncertainty produced by noisy labels. We further define
probability variance of joint tagging probabilities to estimate inner-model
parametric uncertainty, which is used to select and build up new reliable
training instances for the next iteration. Experimental results on two large
datasets reveal that our approach outperforms existing strong baselines and
related methods.
| [
{
"version": "v1",
"created": "Fri, 5 May 2023 20:06:11 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Jun 2023 22:22:24 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Li",
"Yufei",
""
],
[
"Yu",
"Xiao",
""
],
[
"Liu",
"Yanchi",
""
],
[
"Chen",
"Haifeng",
""
],
[
"Liu",
"Cong",
""
]
] | TITLE: Uncertainty-Aware Bootstrap Learning for Joint Extraction on
Distantly-Supervised Data
ABSTRACT: Jointly extracting entity pairs and their relations is challenging when
working on distantly-supervised data with ambiguous or noisy labels. To
mitigate such impact, we propose uncertainty-aware bootstrap learning, which is
motivated by the intuition that the higher uncertainty of an instance, the more
likely the model confidence is inconsistent with the ground truths.
Specifically, we first explore instance-level data uncertainty to create an
initial high-confident examples. Such subset serves as filtering noisy
instances and facilitating the model to converge fast at the early stage.
During bootstrap learning, we propose self-ensembling as a regularizer to
alleviate inter-model uncertainty produced by noisy labels. We further define
probability variance of joint tagging probabilities to estimate inner-model
parametric uncertainty, which is used to select and build up new reliable
training instances for the next iteration. Experimental results on two large
datasets reveal that our approach outperforms existing strong baselines and
related methods.
|
2305.04501 | Junran Wu | Junran Wu, Xueyuan Chen, Bowen Shi, Shangzhe Li, Ke Xu | SEGA: Structural Entropy Guided Anchor View for Graph Contrastive
Learning | ICML'23 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In contrastive learning, the choice of ``view'' controls the information that
the representation captures and influences the performance of the model.
However, leading graph contrastive learning methods generally produce views via
random corruption or learning, which could lead to the loss of essential
information and alteration of semantic information. An anchor view that
maintains the essential information of input graphs for contrastive learning
has been hardly investigated. In this paper, based on the theory of graph
information bottleneck, we deduce the definition of this anchor view; put
differently, \textit{the anchor view with essential information of input graph
is supposed to have the minimal structural uncertainty}. Furthermore, guided by
structural entropy, we implement the anchor view, termed \textbf{SEGA}, for
graph contrastive learning. We extensively validate the proposed anchor view on
various benchmarks regarding graph classification under unsupervised,
semi-supervised, and transfer learning and achieve significant performance
boosts compared to the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 06:52:02 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 08:57:49 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wu",
"Junran",
""
],
[
"Chen",
"Xueyuan",
""
],
[
"Shi",
"Bowen",
""
],
[
"Li",
"Shangzhe",
""
],
[
"Xu",
"Ke",
""
]
] | TITLE: SEGA: Structural Entropy Guided Anchor View for Graph Contrastive
Learning
ABSTRACT: In contrastive learning, the choice of ``view'' controls the information that
the representation captures and influences the performance of the model.
However, leading graph contrastive learning methods generally produce views via
random corruption or learning, which could lead to the loss of essential
information and alteration of semantic information. An anchor view that
maintains the essential information of input graphs for contrastive learning
has been hardly investigated. In this paper, based on the theory of graph
information bottleneck, we deduce the definition of this anchor view; put
differently, \textit{the anchor view with essential information of input graph
is supposed to have the minimal structural uncertainty}. Furthermore, guided by
structural entropy, we implement the anchor view, termed \textbf{SEGA}, for
graph contrastive learning. We extensively validate the proposed anchor view on
various benchmarks regarding graph classification under unsupervised,
semi-supervised, and transfer learning and achieve significant performance
boosts compared to the state-of-the-art methods.
|
2305.05301 | Clementin Boittiaux | Cl\'ementin Boittiaux (IFREMER, COSMER, DYNI), Claire Dune (COSMER),
Maxime Ferrera (IFREMER), Aur\'elien Arnaubec (IFREMER), Ricard Marxer
(DYNI), Marjolaine Matabos (BEEP), Lo\"ic Van Audenhaege (BEEP), Vincent
Hugel (COSMER) | Eiffel Tower: A Deep-Sea Underwater Dataset for Long-Term Visual
Localization | The International Journal of Robotics Research, In press | null | 10.1177/02783649231177322 | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual localization plays an important role in the positioning and navigation
of robotics systems within previously visited environments. When visits occur
over long periods of time, changes in the environment related to seasons or
day-night cycles present a major challenge. Under water, the sources of
variability are due to other factors such as water conditions or growth of
marine organisms. Yet it remains a major obstacle and a much less studied one,
partly due to the lack of data. This paper presents a new deep-sea dataset to
benchmark underwater long-term visual localization. The dataset is composed of
images from four visits to the same hydrothermal vent edifice over the course
of five years. Camera poses and a common geometry of the scene were estimated
using navigation data and Structure-from-Motion. This serves as a reference
when evaluating visual localization techniques. An analysis of the data
provides insights about the major changes observed throughout the years.
Furthermore, several well-established visual localization methods are evaluated
on the dataset, showing there is still room for improvement in underwater
long-term visual localization. The data is made publicly available at
https://www.seanoe.org/data/00810/92226/.
| [
{
"version": "v1",
"created": "Tue, 9 May 2023 09:43:27 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Boittiaux",
"Clémentin",
"",
"IFREMER, COSMER, DYNI"
],
[
"Dune",
"Claire",
"",
"COSMER"
],
[
"Ferrera",
"Maxime",
"",
"IFREMER"
],
[
"Arnaubec",
"Aurélien",
"",
"IFREMER"
],
[
"Marxer",
"Ricard",
"",
"DYNI"
],
[
"Matabos",
"Marjolaine",
"",
"BEEP"
],
[
"Van Audenhaege",
"Loïc",
"",
"BEEP"
],
[
"Hugel",
"Vincent",
"",
"COSMER"
]
] | TITLE: Eiffel Tower: A Deep-Sea Underwater Dataset for Long-Term Visual
Localization
ABSTRACT: Visual localization plays an important role in the positioning and navigation
of robotics systems within previously visited environments. When visits occur
over long periods of time, changes in the environment related to seasons or
day-night cycles present a major challenge. Under water, the sources of
variability are due to other factors such as water conditions or growth of
marine organisms. Yet it remains a major obstacle and a much less studied one,
partly due to the lack of data. This paper presents a new deep-sea dataset to
benchmark underwater long-term visual localization. The dataset is composed of
images from four visits to the same hydrothermal vent edifice over the course
of five years. Camera poses and a common geometry of the scene were estimated
using navigation data and Structure-from-Motion. This serves as a reference
when evaluating visual localization techniques. An analysis of the data
provides insights about the major changes observed throughout the years.
Furthermore, several well-established visual localization methods are evaluated
on the dataset, showing there is still room for improvement in underwater
long-term visual localization. The data is made publicly available at
https://www.seanoe.org/data/00810/92226/.
|
2305.07829 | Jun Cheng | Jun Cheng, Honglei Su, Jari Korhonen | No-Reference Point Cloud Quality Assessment via Weighted Patch Quality
Prediction | 6 pages, 5 figures, Accepted by International Conference on Software
Engineering and Knowledge Engineering(SEKE2023) | null | 10.18293/SEKE2023-185 | null | cs.CV cs.LG eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the rapid development of 3D vision applications based on point clouds,
point cloud quality assessment(PCQA) is becoming an important research topic.
However, the prior PCQA methods ignore the effect of local quality variance
across different areas of the point cloud. To take an advantage of the quality
distribution imbalance, we propose a no-reference point cloud quality
assessment (NR-PCQA) method with local area correlation analysis capability,
denoted as COPP-Net. More specifically, we split a point cloud into patches,
generate texture and structure features for each patch, and fuse them into
patch features to predict patch quality. Then, we gather the features of all
the patches of a point cloud for correlation analysis, to obtain the
correlation weights. Finally, the predicted qualities and correlation weights
for all the patches are used to derive the final quality score. Experimental
results show that our method outperforms the state-of-the-art benchmark NR-PCQA
methods. The source code for the proposed COPP-Net can be found at
https://github.com/philox12358/COPP-Net.
| [
{
"version": "v1",
"created": "Sat, 13 May 2023 03:20:33 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 09:27:33 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Cheng",
"Jun",
""
],
[
"Su",
"Honglei",
""
],
[
"Korhonen",
"Jari",
""
]
] | TITLE: No-Reference Point Cloud Quality Assessment via Weighted Patch Quality
Prediction
ABSTRACT: With the rapid development of 3D vision applications based on point clouds,
point cloud quality assessment(PCQA) is becoming an important research topic.
However, the prior PCQA methods ignore the effect of local quality variance
across different areas of the point cloud. To take an advantage of the quality
distribution imbalance, we propose a no-reference point cloud quality
assessment (NR-PCQA) method with local area correlation analysis capability,
denoted as COPP-Net. More specifically, we split a point cloud into patches,
generate texture and structure features for each patch, and fuse them into
patch features to predict patch quality. Then, we gather the features of all
the patches of a point cloud for correlation analysis, to obtain the
correlation weights. Finally, the predicted qualities and correlation weights
for all the patches are used to derive the final quality score. Experimental
results show that our method outperforms the state-of-the-art benchmark NR-PCQA
methods. The source code for the proposed COPP-Net can be found at
https://github.com/philox12358/COPP-Net.
|
2305.15182 | Junran Wu | He Zhu, Chong Zhang, Junjie Huang, Junran Wu, Ke Xu | HiTIN: Hierarchy-aware Tree Isomorphism Network for Hierarchical Text
Classification | Accepted by ACL'23 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Hierarchical text classification (HTC) is a challenging subtask of
multi-label classification as the labels form a complex hierarchical structure.
Existing dual-encoder methods in HTC achieve weak performance gains with huge
memory overheads and their structure encoders heavily rely on domain knowledge.
Under such observation, we tend to investigate the feasibility of a
memory-friendly model with strong generalization capability that could boost
the performance of HTC without prior statistics or label semantics. In this
paper, we propose Hierarchy-aware Tree Isomorphism Network (HiTIN) to enhance
the text representations with only syntactic information of the label
hierarchy. Specifically, we convert the label hierarchy into an unweighted tree
structure, termed coding tree, with the guidance of structural entropy. Then we
design a structure encoder to incorporate hierarchy-aware information in the
coding tree into text representations. Besides the text encoder, HiTIN only
contains a few multi-layer perceptions and linear transformations, which
greatly saves memory. We conduct experiments on three commonly used datasets
and the results demonstrate that HiTIN could achieve better test performance
and less memory consumption than state-of-the-art (SOTA) methods.
| [
{
"version": "v1",
"created": "Wed, 24 May 2023 14:14:08 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 08:53:14 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Zhu",
"He",
""
],
[
"Zhang",
"Chong",
""
],
[
"Huang",
"Junjie",
""
],
[
"Wu",
"Junran",
""
],
[
"Xu",
"Ke",
""
]
] | TITLE: HiTIN: Hierarchy-aware Tree Isomorphism Network for Hierarchical Text
Classification
ABSTRACT: Hierarchical text classification (HTC) is a challenging subtask of
multi-label classification as the labels form a complex hierarchical structure.
Existing dual-encoder methods in HTC achieve weak performance gains with huge
memory overheads and their structure encoders heavily rely on domain knowledge.
Under such observation, we tend to investigate the feasibility of a
memory-friendly model with strong generalization capability that could boost
the performance of HTC without prior statistics or label semantics. In this
paper, we propose Hierarchy-aware Tree Isomorphism Network (HiTIN) to enhance
the text representations with only syntactic information of the label
hierarchy. Specifically, we convert the label hierarchy into an unweighted tree
structure, termed coding tree, with the guidance of structural entropy. Then we
design a structure encoder to incorporate hierarchy-aware information in the
coding tree into text representations. Besides the text encoder, HiTIN only
contains a few multi-layer perceptions and linear transformations, which
greatly saves memory. We conduct experiments on three commonly used datasets
and the results demonstrate that HiTIN could achieve better test performance
and less memory consumption than state-of-the-art (SOTA) methods.
|
2305.16255 | Paul Ghelasi | Paul Ghelasi, Florian Ziel | Hierarchical forecasting for aggregated curves with an application to
day-ahead electricity price auctions | 34 pages, 6 figures. International Journal of Forecasting (2022) | null | 10.1016/j.ijforecast.2022.11.004 | null | stat.AP cs.LG econ.EM q-fin.TR stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Aggregated curves are common structures in economics and finance, and the
most prominent examples are supply and demand curves. In this study, we exploit
the fact that all aggregated curves have an intrinsic hierarchical structure,
and thus hierarchical reconciliation methods can be used to improve the
forecast accuracy. We provide an in-depth theory on how aggregated curves can
be constructed or deconstructed, and conclude that these methods are equivalent
under weak assumptions. We consider multiple reconciliation methods for
aggregated curves, including previously established bottom-up, top-down, and
linear optimal reconciliation approaches. We also present a new benchmark
reconciliation method called 'aggregated-down' with similar complexity to
bottom-up and top-down approaches, but it tends to provide better accuracy in
this setup. We conducted an empirical forecasting study on the German day-ahead
power auction market by predicting the demand and supply curves, where their
equilibrium determines the electricity price for the next day. Our results
demonstrate that hierarchical reconciliation methods can be used to improve the
forecasting accuracy of aggregated curves.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 17:10:54 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Ghelasi",
"Paul",
""
],
[
"Ziel",
"Florian",
""
]
] | TITLE: Hierarchical forecasting for aggregated curves with an application to
day-ahead electricity price auctions
ABSTRACT: Aggregated curves are common structures in economics and finance, and the
most prominent examples are supply and demand curves. In this study, we exploit
the fact that all aggregated curves have an intrinsic hierarchical structure,
and thus hierarchical reconciliation methods can be used to improve the
forecast accuracy. We provide an in-depth theory on how aggregated curves can
be constructed or deconstructed, and conclude that these methods are equivalent
under weak assumptions. We consider multiple reconciliation methods for
aggregated curves, including previously established bottom-up, top-down, and
linear optimal reconciliation approaches. We also present a new benchmark
reconciliation method called 'aggregated-down' with similar complexity to
bottom-up and top-down approaches, but it tends to provide better accuracy in
this setup. We conducted an empirical forecasting study on the German day-ahead
power auction market by predicting the demand and supply curves, where their
equilibrium determines the electricity price for the next day. Our results
demonstrate that hierarchical reconciliation methods can be used to improve the
forecasting accuracy of aggregated curves.
|
2305.18170 | Zhanming Jie | Zhanming Jie, Wei Lu | Leveraging Training Data in Few-Shot Prompting for Numerical Reasoning | ACL 2023 Findings | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Chain-of-thought (CoT) prompting with large language models has proven
effective in numerous natural language processing tasks, but designing prompts
that generalize well to diverse problem types can be challenging, especially in
the context of math word problem (MWP) solving. Additionally, it is common to
have a large amount of training data that have a better diversity coverage but
CoT annotations are not available, which limits the use of supervised learning
techniques. To address these issues, we investigate two approaches to leverage
the training data in a few-shot prompting scenario: dynamic program prompting
and program distillation. Our approach is largely inspired by Gao et al.,
(2022), where they proposed to replace the CoT with the programs as the
intermediate reasoning step. Such a prompting strategy allows us to accurately
verify the answer correctness through program execution in MWP solving. Our
dynamic program prompting involves annotating the training data by sampling
correct programs from a large language model, while program distillation
involves adapting a smaller model to the program-annotated training data. Our
experiments on three standard MWP datasets demonstrate the effectiveness of
these approaches, yielding significant improvements over previous baselines for
prompting and fine-tuning. Our results suggest that leveraging a large amount
of training data can improve the generalization ability of prompts and boost
the performance of fine-tuned small models in MWP solving.
| [
{
"version": "v1",
"created": "Mon, 29 May 2023 16:01:40 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 05:00:30 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Jie",
"Zhanming",
""
],
[
"Lu",
"Wei",
""
]
] | TITLE: Leveraging Training Data in Few-Shot Prompting for Numerical Reasoning
ABSTRACT: Chain-of-thought (CoT) prompting with large language models has proven
effective in numerous natural language processing tasks, but designing prompts
that generalize well to diverse problem types can be challenging, especially in
the context of math word problem (MWP) solving. Additionally, it is common to
have a large amount of training data that have a better diversity coverage but
CoT annotations are not available, which limits the use of supervised learning
techniques. To address these issues, we investigate two approaches to leverage
the training data in a few-shot prompting scenario: dynamic program prompting
and program distillation. Our approach is largely inspired by Gao et al.,
(2022), where they proposed to replace the CoT with the programs as the
intermediate reasoning step. Such a prompting strategy allows us to accurately
verify the answer correctness through program execution in MWP solving. Our
dynamic program prompting involves annotating the training data by sampling
correct programs from a large language model, while program distillation
involves adapting a smaller model to the program-annotated training data. Our
experiments on three standard MWP datasets demonstrate the effectiveness of
these approaches, yielding significant improvements over previous baselines for
prompting and fine-tuning. Our results suggest that leveraging a large amount
of training data can improve the generalization ability of prompts and boost
the performance of fine-tuned small models in MWP solving.
|
2305.19590 | Jiahui Huang | Jiahui Huang, Zan Gojcic, Matan Atzmon, Or Litany, Sanja Fidler,
Francis Williams | Neural Kernel Surface Reconstruction | CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | We present a novel method for reconstructing a 3D implicit surface from a
large-scale, sparse, and noisy point cloud. Our approach builds upon the
recently introduced Neural Kernel Fields (NKF) representation. It enjoys
similar generalization capabilities to NKF, while simultaneously addressing its
main limitations: (a) We can scale to large scenes through compactly supported
kernel functions, which enable the use of memory-efficient sparse linear
solvers. (b) We are robust to noise, through a gradient fitting solve. (c) We
minimize training requirements, enabling us to learn from any dataset of dense
oriented points, and even mix training data consisting of objects and scenes at
different scales. Our method is capable of reconstructing millions of points in
a few seconds, and handling very large scenes in an out-of-core fashion. We
achieve state-of-the-art results on reconstruction benchmarks consisting of
single objects, indoor scenes, and outdoor scenes.
| [
{
"version": "v1",
"created": "Wed, 31 May 2023 06:25:18 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 09:55:13 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Huang",
"Jiahui",
""
],
[
"Gojcic",
"Zan",
""
],
[
"Atzmon",
"Matan",
""
],
[
"Litany",
"Or",
""
],
[
"Fidler",
"Sanja",
""
],
[
"Williams",
"Francis",
""
]
] | TITLE: Neural Kernel Surface Reconstruction
ABSTRACT: We present a novel method for reconstructing a 3D implicit surface from a
large-scale, sparse, and noisy point cloud. Our approach builds upon the
recently introduced Neural Kernel Fields (NKF) representation. It enjoys
similar generalization capabilities to NKF, while simultaneously addressing its
main limitations: (a) We can scale to large scenes through compactly supported
kernel functions, which enable the use of memory-efficient sparse linear
solvers. (b) We are robust to noise, through a gradient fitting solve. (c) We
minimize training requirements, enabling us to learn from any dataset of dense
oriented points, and even mix training data consisting of objects and scenes at
different scales. Our method is capable of reconstructing millions of points in
a few seconds, and handling very large scenes in an out-of-core fashion. We
achieve state-of-the-art results on reconstruction benchmarks consisting of
single objects, indoor scenes, and outdoor scenes.
|
2306.01985 | Xuhui Zhou | Xuhui Zhou, Hao Zhu, Akhila Yerukola, and Thomas Davidson, Jena D.
Hwang, Swabha Swayamdipta, Maarten Sap | COBRA Frames: Contextual Reasoning about Effects and Harms of Offensive
Statements | Accepted to Findings of ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Warning: This paper contains content that may be offensive or upsetting.
Understanding the harms and offensiveness of statements requires reasoning
about the social and situational context in which statements are made. For
example, the utterance "your English is very good" may implicitly signal an
insult when uttered by a white man to a non-white colleague, but uttered by an
ESL teacher to their student would be interpreted as a genuine compliment. Such
contextual factors have been largely ignored by previous approaches to toxic
language detection. We introduce COBRA frames, the first context-aware
formalism for explaining the intents, reactions, and harms of offensive or
biased statements grounded in their social and situational context. We create
COBRACORPUS, a dataset of 33k potentially offensive statements paired with
machine-generated contexts and free-text explanations of offensiveness, implied
biases, speaker intents, and listener reactions. To study the contextual
dynamics of offensiveness, we train models to generate COBRA explanations, with
and without access to the context. We find that explanations by
context-agnostic models are significantly worse than by context-aware ones,
especially in situations where the context inverts the statement's
offensiveness (29% accuracy drop). Our work highlights the importance and
feasibility of contextualized NLP by modeling social factors.
| [
{
"version": "v1",
"created": "Sat, 3 Jun 2023 02:47:24 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 01:49:06 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Zhou",
"Xuhui",
""
],
[
"Zhu",
"Hao",
""
],
[
"Yerukola",
"Akhila",
""
],
[
"Davidson",
"Thomas",
""
],
[
"Hwang",
"Jena D.",
""
],
[
"Swayamdipta",
"Swabha",
""
],
[
"Sap",
"Maarten",
""
]
] | TITLE: COBRA Frames: Contextual Reasoning about Effects and Harms of Offensive
Statements
ABSTRACT: Warning: This paper contains content that may be offensive or upsetting.
Understanding the harms and offensiveness of statements requires reasoning
about the social and situational context in which statements are made. For
example, the utterance "your English is very good" may implicitly signal an
insult when uttered by a white man to a non-white colleague, but uttered by an
ESL teacher to their student would be interpreted as a genuine compliment. Such
contextual factors have been largely ignored by previous approaches to toxic
language detection. We introduce COBRA frames, the first context-aware
formalism for explaining the intents, reactions, and harms of offensive or
biased statements grounded in their social and situational context. We create
COBRACORPUS, a dataset of 33k potentially offensive statements paired with
machine-generated contexts and free-text explanations of offensiveness, implied
biases, speaker intents, and listener reactions. To study the contextual
dynamics of offensiveness, we train models to generate COBRA explanations, with
and without access to the context. We find that explanations by
context-agnostic models are significantly worse than by context-aware ones,
especially in situations where the context inverts the statement's
offensiveness (29% accuracy drop). Our work highlights the importance and
feasibility of contextualized NLP by modeling social factors.
|
2306.02077 | Georgios Peikos | Georgios Peikos, Symeon Symeonidis, Pranav Kasela, Gabriella Pasi | Utilizing ChatGPT to Enhance Clinical Trial Enrollment | Under Review | null | null | null | cs.IR cs.CL | http://creativecommons.org/licenses/by/4.0/ | Clinical trials are a critical component of evaluating the effectiveness of
new medical interventions and driving advancements in medical research.
Therefore, timely enrollment of patients is crucial to prevent delays or
premature termination of trials. In this context, Electronic Health Records
(EHRs) have emerged as a valuable tool for identifying and enrolling eligible
participants. In this study, we propose an automated approach that leverages
ChatGPT, a large language model, to extract patient-related information from
unstructured clinical notes and generate search queries for retrieving
potentially eligible clinical trials. Our empirical evaluation, conducted on
two benchmark retrieval collections, shows improved retrieval performance
compared to existing approaches when several general-purposed and task-specific
prompts are used. Notably, ChatGPT-generated queries also outperform
human-generated queries in terms of retrieval performance. These findings
highlight the potential use of ChatGPT to enhance clinical trial enrollment
while ensuring the quality of medical service and minimizing direct risks to
patients.
| [
{
"version": "v1",
"created": "Sat, 3 Jun 2023 10:54:23 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Peikos",
"Georgios",
""
],
[
"Symeonidis",
"Symeon",
""
],
[
"Kasela",
"Pranav",
""
],
[
"Pasi",
"Gabriella",
""
]
] | TITLE: Utilizing ChatGPT to Enhance Clinical Trial Enrollment
ABSTRACT: Clinical trials are a critical component of evaluating the effectiveness of
new medical interventions and driving advancements in medical research.
Therefore, timely enrollment of patients is crucial to prevent delays or
premature termination of trials. In this context, Electronic Health Records
(EHRs) have emerged as a valuable tool for identifying and enrolling eligible
participants. In this study, we propose an automated approach that leverages
ChatGPT, a large language model, to extract patient-related information from
unstructured clinical notes and generate search queries for retrieving
potentially eligible clinical trials. Our empirical evaluation, conducted on
two benchmark retrieval collections, shows improved retrieval performance
compared to existing approaches when several general-purposed and task-specific
prompts are used. Notably, ChatGPT-generated queries also outperform
human-generated queries in terms of retrieval performance. These findings
highlight the potential use of ChatGPT to enhance clinical trial enrollment
while ensuring the quality of medical service and minimizing direct risks to
patients.
|
2306.02133 | Sushovan Majhi | Sushovan Majhi | Graph Mover's Distance: An Efficiently Computable Distance Measure for
Geometric Graphs | arXiv admin note: text overlap with arXiv:2209.12869 | null | null | null | cs.CG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Many applications in pattern recognition represent patterns as a geometric
graph. The geometric graph distance (GGD) has recently been studied as a
meaningful measure of similarity between two geometric graphs. Since computing
the GGD is known to be $\mathcal{NP}$-hard, the distance measure proves an
impractical choice for applications. As a computationally tractable
alternative, we propose in this paper the Graph Mover's Distance (GMD), which
has been formulated as an instance of the earth mover's distance. The
computation of the GMD between two geometric graphs with at most $n$ vertices
takes only $O(n^3)$-time. Alongside studying the metric properties of the GMD,
we investigate the stability of the GGD and GMD. The GMD also demonstrates
extremely promising empirical evidence at recognizing letter drawings from the
{\tt LETTER} dataset \cite{da_vitoria_lobo_iam_2008}.
| [
{
"version": "v1",
"created": "Sat, 3 Jun 2023 15:06:12 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Majhi",
"Sushovan",
""
]
] | TITLE: Graph Mover's Distance: An Efficiently Computable Distance Measure for
Geometric Graphs
ABSTRACT: Many applications in pattern recognition represent patterns as a geometric
graph. The geometric graph distance (GGD) has recently been studied as a
meaningful measure of similarity between two geometric graphs. Since computing
the GGD is known to be $\mathcal{NP}$-hard, the distance measure proves an
impractical choice for applications. As a computationally tractable
alternative, we propose in this paper the Graph Mover's Distance (GMD), which
has been formulated as an instance of the earth mover's distance. The
computation of the GMD between two geometric graphs with at most $n$ vertices
takes only $O(n^3)$-time. Alongside studying the metric properties of the GMD,
we investigate the stability of the GGD and GMD. The GMD also demonstrates
extremely promising empirical evidence at recognizing letter drawings from the
{\tt LETTER} dataset \cite{da_vitoria_lobo_iam_2008}.
|
2306.02140 | Qingxin Xia | Qingxin Xia and Takuya Maekawa and Takahiro Hara | Unsupervised Human Activity Recognition through Two-stage Prompting with
ChatGPT | 4 pages | null | null | null | cs.HC cs.CL | http://creativecommons.org/licenses/by/4.0/ | Wearable sensor devices, which offer the advantage of recording daily objects
used by a person while performing an activity, enable the feasibility of
unsupervised Human Activity Recognition (HAR). Unfortunately, previous
unsupervised approaches using the usage sequence of objects usually require a
proper description of activities manually prepared by humans. Instead, we
leverage the knowledge embedded in a Large Language Model (LLM) of ChatGPT.
Because the sequence of objects robustly characterizes the activity identity,
it is possible that ChatGPT already learned the association between activities
and objects from existing contexts. However, previous prompt engineering for
ChatGPT exhibits limited generalization ability when dealing with a list of
words (i.e., sequence of objects) due to the similar weighting assigned to each
word in the list. In this study, we propose a two-stage prompt engineering,
which first guides ChatGPT to generate activity descriptions associated with
objects while emphasizing important objects for distinguishing similar
activities; then outputs activity classes and explanations for enhancing the
contexts that are helpful for HAR. To the best of our knowledge, this is the
first study that utilizes ChatGPT to recognize activities using objects in an
unsupervised manner. We conducted our approach on three datasets and
demonstrated the state-of-the-art performance.
| [
{
"version": "v1",
"created": "Sat, 3 Jun 2023 15:41:59 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Xia",
"Qingxin",
""
],
[
"Maekawa",
"Takuya",
""
],
[
"Hara",
"Takahiro",
""
]
] | TITLE: Unsupervised Human Activity Recognition through Two-stage Prompting with
ChatGPT
ABSTRACT: Wearable sensor devices, which offer the advantage of recording daily objects
used by a person while performing an activity, enable the feasibility of
unsupervised Human Activity Recognition (HAR). Unfortunately, previous
unsupervised approaches using the usage sequence of objects usually require a
proper description of activities manually prepared by humans. Instead, we
leverage the knowledge embedded in a Large Language Model (LLM) of ChatGPT.
Because the sequence of objects robustly characterizes the activity identity,
it is possible that ChatGPT already learned the association between activities
and objects from existing contexts. However, previous prompt engineering for
ChatGPT exhibits limited generalization ability when dealing with a list of
words (i.e., sequence of objects) due to the similar weighting assigned to each
word in the list. In this study, we propose a two-stage prompt engineering,
which first guides ChatGPT to generate activity descriptions associated with
objects while emphasizing important objects for distinguishing similar
activities; then outputs activity classes and explanations for enhancing the
contexts that are helpful for HAR. To the best of our knowledge, this is the
first study that utilizes ChatGPT to recognize activities using objects in an
unsupervised manner. We conducted our approach on three datasets and
demonstrated the state-of-the-art performance.
|
2306.02175 | Shuo Lei | Shuo Lei, Xuchao Zhang, Jianfeng He, Fanglan Chen, Chang-Tien Lu | TART: Improved Few-shot Text Classification Using Task-Adaptive
Reference Transformation | 11 pages, 5 figures. Accepted by ACL 2023. arXiv admin note: text
overlap with arXiv:2107.12262 by other authors | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Meta-learning has emerged as a trending technique to tackle few-shot text
classification and achieve state-of-the-art performance. However, the
performance of existing approaches heavily depends on the inter-class variance
of the support set. As a result, it can perform well on tasks when the
semantics of sampled classes are distinct while failing to differentiate
classes with similar semantics. In this paper, we propose a novel Task-Adaptive
Reference Transformation (TART) network, aiming to enhance the generalization
by transforming the class prototypes to per-class fixed reference points in
task-adaptive metric spaces. To further maximize divergence between transformed
prototypes in task-adaptive metric spaces, TART introduces a discriminative
reference regularization among transformed prototypes. Extensive experiments
are conducted on four benchmark datasets and our method demonstrates clear
superiority over the state-of-the-art models in all the datasets. In
particular, our model surpasses the state-of-the-art method by 7.4% and 5.4% in
1-shot and 5-shot classification on the 20 Newsgroups dataset, respectively.
| [
{
"version": "v1",
"created": "Sat, 3 Jun 2023 18:38:02 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lei",
"Shuo",
""
],
[
"Zhang",
"Xuchao",
""
],
[
"He",
"Jianfeng",
""
],
[
"Chen",
"Fanglan",
""
],
[
"Lu",
"Chang-Tien",
""
]
] | TITLE: TART: Improved Few-shot Text Classification Using Task-Adaptive
Reference Transformation
ABSTRACT: Meta-learning has emerged as a trending technique to tackle few-shot text
classification and achieve state-of-the-art performance. However, the
performance of existing approaches heavily depends on the inter-class variance
of the support set. As a result, it can perform well on tasks when the
semantics of sampled classes are distinct while failing to differentiate
classes with similar semantics. In this paper, we propose a novel Task-Adaptive
Reference Transformation (TART) network, aiming to enhance the generalization
by transforming the class prototypes to per-class fixed reference points in
task-adaptive metric spaces. To further maximize divergence between transformed
prototypes in task-adaptive metric spaces, TART introduces a discriminative
reference regularization among transformed prototypes. Extensive experiments
are conducted on four benchmark datasets and our method demonstrates clear
superiority over the state-of-the-art models in all the datasets. In
particular, our model surpasses the state-of-the-art method by 7.4% and 5.4% in
1-shot and 5-shot classification on the 20 Newsgroups dataset, respectively.
|
2306.02239 | Shuchang Liu | Shuchang Liu, Qingpeng Cai, Zhankui He, Bowen Sun, Julian McAuley,
Dong Zheng, Peng Jiang, and Kun Gai | Generative Flow Network for Listwise Recommendation | 11 pages, 5 figures, 7 tables | KDD '23, August 6-10, 2023, Long Beach, CA, USA | 10.1145/3580305.3599364 | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Personalized recommender systems fulfill the daily demands of customers and
boost online businesses. The goal is to learn a policy that can generate a list
of items that matches the user's demand or interest. While most existing
methods learn a pointwise scoring model that predicts the ranking score of each
individual item, recent research shows that the listwise approach can further
improve the recommendation quality by modeling the intra-list correlations of
items that are exposed together. This has motivated the recent list reranking
and generative recommendation approaches that optimize the overall utility of
the entire list. However, it is challenging to explore the combinatorial space
of list actions and existing methods that use cross-entropy loss may suffer
from low diversity issues. In this work, we aim to learn a policy that can
generate sufficiently diverse item lists for users while maintaining high
recommendation quality. The proposed solution, GFN4Rec, is a generative method
that takes the insight of the flow network to ensure the alignment between list
generation probability and its reward. The key advantages of our solution are
the log scale reward matching loss that intrinsically improves the generation
diversity and the autoregressive item selection model that captures the item
mutual influences while capturing future reward of the list. As validation of
our method's effectiveness and its superior diversity during active
exploration, we conduct experiments on simulated online environments as well as
an offline evaluation framework for two real-world datasets.
| [
{
"version": "v1",
"created": "Sun, 4 Jun 2023 02:50:11 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 06:14:05 GMT"
},
{
"version": "v3",
"created": "Thu, 8 Jun 2023 07:49:42 GMT"
},
{
"version": "v4",
"created": "Fri, 9 Jun 2023 07:03:01 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Liu",
"Shuchang",
""
],
[
"Cai",
"Qingpeng",
""
],
[
"He",
"Zhankui",
""
],
[
"Sun",
"Bowen",
""
],
[
"McAuley",
"Julian",
""
],
[
"Zheng",
"Dong",
""
],
[
"Jiang",
"Peng",
""
],
[
"Gai",
"Kun",
""
]
] | TITLE: Generative Flow Network for Listwise Recommendation
ABSTRACT: Personalized recommender systems fulfill the daily demands of customers and
boost online businesses. The goal is to learn a policy that can generate a list
of items that matches the user's demand or interest. While most existing
methods learn a pointwise scoring model that predicts the ranking score of each
individual item, recent research shows that the listwise approach can further
improve the recommendation quality by modeling the intra-list correlations of
items that are exposed together. This has motivated the recent list reranking
and generative recommendation approaches that optimize the overall utility of
the entire list. However, it is challenging to explore the combinatorial space
of list actions and existing methods that use cross-entropy loss may suffer
from low diversity issues. In this work, we aim to learn a policy that can
generate sufficiently diverse item lists for users while maintaining high
recommendation quality. The proposed solution, GFN4Rec, is a generative method
that takes the insight of the flow network to ensure the alignment between list
generation probability and its reward. The key advantages of our solution are
the log scale reward matching loss that intrinsically improves the generation
diversity and the autoregressive item selection model that captures the item
mutual influences while capturing future reward of the list. As validation of
our method's effectiveness and its superior diversity during active
exploration, we conduct experiments on simulated online environments as well as
an offline evaluation framework for two real-world datasets.
|
2306.02693 | Hyunsoo Cho | Hyunsoo Cho, Youna Kim, Sang-goo Lee | CELDA: Leveraging Black-box Language Model as Enhanced Classifier
without Labels | ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Utilizing language models (LMs) without internal access is becoming an
attractive paradigm in the field of NLP as many cutting-edge LMs are released
through APIs and boast a massive scale. The de-facto method in this type of
black-box scenario is known as prompting, which has shown progressive
performance enhancements in situations where data labels are scarce or
unavailable. Despite their efficacy, they still fall short in comparison to
fully supervised counterparts and are generally brittle to slight
modifications. In this paper, we propose Clustering-enhanced Linear
Discriminative Analysis, a novel approach that improves the text classification
accuracy with a very weak-supervision signal (i.e., name of the labels). Our
framework draws a precise decision boundary without accessing weights or
gradients of the LM model or data labels. The core ideas of CELDA are twofold:
(1) extracting a refined pseudo-labeled dataset from an unlabeled dataset, and
(2) training a lightweight and robust model on the top of LM, which learns an
accurate decision boundary from an extracted noisy dataset. Throughout in-depth
investigations on various datasets, we demonstrated that CELDA reaches new
state-of-the-art in weakly-supervised text classification and narrows the gap
with a fully-supervised model. Additionally, our proposed methodology can be
applied universally to any LM and has the potential to scale to larger models,
making it a more viable option for utilizing large LMs.
| [
{
"version": "v1",
"created": "Mon, 5 Jun 2023 08:35:31 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 05:16:21 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Cho",
"Hyunsoo",
""
],
[
"Kim",
"Youna",
""
],
[
"Lee",
"Sang-goo",
""
]
] | TITLE: CELDA: Leveraging Black-box Language Model as Enhanced Classifier
without Labels
ABSTRACT: Utilizing language models (LMs) without internal access is becoming an
attractive paradigm in the field of NLP as many cutting-edge LMs are released
through APIs and boast a massive scale. The de-facto method in this type of
black-box scenario is known as prompting, which has shown progressive
performance enhancements in situations where data labels are scarce or
unavailable. Despite their efficacy, they still fall short in comparison to
fully supervised counterparts and are generally brittle to slight
modifications. In this paper, we propose Clustering-enhanced Linear
Discriminative Analysis, a novel approach that improves the text classification
accuracy with a very weak-supervision signal (i.e., name of the labels). Our
framework draws a precise decision boundary without accessing weights or
gradients of the LM model or data labels. The core ideas of CELDA are twofold:
(1) extracting a refined pseudo-labeled dataset from an unlabeled dataset, and
(2) training a lightweight and robust model on the top of LM, which learns an
accurate decision boundary from an extracted noisy dataset. Throughout in-depth
investigations on various datasets, we demonstrated that CELDA reaches new
state-of-the-art in weakly-supervised text classification and narrows the gap
with a fully-supervised model. Additionally, our proposed methodology can be
applied universally to any LM and has the potential to scale to larger models,
making it a more viable option for utilizing large LMs.
|
2306.02838 | Veronica Lachi | Veronica Lachi and Giovanna Maria Dimitri and Alessandro Di Stefano
and Pietro Li\`o and Monica Bianchini and Chiara Mocenni | Impact of the Covid 19 outbreaks on the italian twitter vaccination
debat: a network based analysis | null | null | null | null | cs.SI math.DS physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | Vaccine hesitancy, or the reluctance to be vaccinated, is a phenomenon that
has recently become particularly significant, in conjunction with the
vaccination campaign against COVID-19. During the lockdown period, necessary to
control the spread of the virus, social networks have played an important role
in the Italian debate on vaccination, generally representing the easiest and
safest way to exchange opinions and maintain some form of sociability. Among
social network platforms, Twitter has assumed a strategic role in driving the
public opinion, creating compact groups of users sharing similar views towards
the utility, uselessness or even dangerousness of vaccines. In this paper, we
present a new, publicly available, dataset of Italian tweets, TwitterVax,
collected in the period January 2019--May 2022. Considering monthly data,
gathered into forty one retweet networks -- where nodes identify users and
edges are present between users who have retweeted each other -- we performed
community detection within the networks, analyzing their evolution and
polarization with respect to NoVax and ProVax users through time. This allowed
us to clearly discover debate trends as well as identify potential key moments
and actors in opinion flows, characterizing the main features and tweeting
behavior of the two communities.
| [
{
"version": "v1",
"created": "Mon, 5 Jun 2023 12:43:39 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lachi",
"Veronica",
""
],
[
"Dimitri",
"Giovanna Maria",
""
],
[
"Di Stefano",
"Alessandro",
""
],
[
"Liò",
"Pietro",
""
],
[
"Bianchini",
"Monica",
""
],
[
"Mocenni",
"Chiara",
""
]
] | TITLE: Impact of the Covid 19 outbreaks on the italian twitter vaccination
debat: a network based analysis
ABSTRACT: Vaccine hesitancy, or the reluctance to be vaccinated, is a phenomenon that
has recently become particularly significant, in conjunction with the
vaccination campaign against COVID-19. During the lockdown period, necessary to
control the spread of the virus, social networks have played an important role
in the Italian debate on vaccination, generally representing the easiest and
safest way to exchange opinions and maintain some form of sociability. Among
social network platforms, Twitter has assumed a strategic role in driving the
public opinion, creating compact groups of users sharing similar views towards
the utility, uselessness or even dangerousness of vaccines. In this paper, we
present a new, publicly available, dataset of Italian tweets, TwitterVax,
collected in the period January 2019--May 2022. Considering monthly data,
gathered into forty one retweet networks -- where nodes identify users and
edges are present between users who have retweeted each other -- we performed
community detection within the networks, analyzing their evolution and
polarization with respect to NoVax and ProVax users through time. This allowed
us to clearly discover debate trends as well as identify potential key moments
and actors in opinion flows, characterizing the main features and tweeting
behavior of the two communities.
|
2306.03042 | Amin Shoari Nejad | Amin Shoari Nejad, Roc\'io Alaiz-Rodr\'iguez, Gerard D. McCarthy,
Brian Kelleher, Anthony Grey, Andrew Parnell | SERT: A Transfomer Based Model for Spatio-Temporal Sensor Data with
Missing Values for Environmental Monitoring | 11 pages, 7 figures | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Environmental monitoring is crucial to our understanding of climate change,
biodiversity loss and pollution. The availability of large-scale
spatio-temporal data from sources such as sensors and satellites allows us to
develop sophisticated models for forecasting and understanding key drivers.
However, the data collected from sensors often contain missing values due to
faulty equipment or maintenance issues. The missing values rarely occur
simultaneously leading to data that are multivariate misaligned sparse time
series. We propose two models that are capable of performing multivariate
spatio-temporal forecasting while handling missing data naturally without the
need for imputation. The first model is a transformer-based model, which we
name SERT (Spatio-temporal Encoder Representations from Transformers). The
second is a simpler model named SST-ANN (Sparse Spatio-Temporal Artificial
Neural Network) which is capable of providing interpretable results. We conduct
extensive experiments on two different datasets for multivariate
spatio-temporal forecasting and show that our models have competitive or
superior performance to those at the state-of-the-art.
| [
{
"version": "v1",
"created": "Mon, 5 Jun 2023 17:06:23 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 08:26:57 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Nejad",
"Amin Shoari",
""
],
[
"Alaiz-Rodríguez",
"Rocío",
""
],
[
"McCarthy",
"Gerard D.",
""
],
[
"Kelleher",
"Brian",
""
],
[
"Grey",
"Anthony",
""
],
[
"Parnell",
"Andrew",
""
]
] | TITLE: SERT: A Transfomer Based Model for Spatio-Temporal Sensor Data with
Missing Values for Environmental Monitoring
ABSTRACT: Environmental monitoring is crucial to our understanding of climate change,
biodiversity loss and pollution. The availability of large-scale
spatio-temporal data from sources such as sensors and satellites allows us to
develop sophisticated models for forecasting and understanding key drivers.
However, the data collected from sensors often contain missing values due to
faulty equipment or maintenance issues. The missing values rarely occur
simultaneously leading to data that are multivariate misaligned sparse time
series. We propose two models that are capable of performing multivariate
spatio-temporal forecasting while handling missing data naturally without the
need for imputation. The first model is a transformer-based model, which we
name SERT (Spatio-temporal Encoder Representations from Transformers). The
second is a simpler model named SST-ANN (Sparse Spatio-Temporal Artificial
Neural Network) which is capable of providing interpretable results. We conduct
extensive experiments on two different datasets for multivariate
spatio-temporal forecasting and show that our models have competitive or
superior performance to those at the state-of-the-art.
|
2306.03514 | Xinyu Huang | Youcai Zhang, Xinyu Huang, Jinyu Ma, Zhaoyang Li, Zhaochuan Luo,
Yanchun Xie, Yuzhuo Qin, Tong Luo, Yaqian Li, Shilong Liu, Yandong Guo, Lei
Zhang | Recognize Anything: A Strong Image Tagging Model | Homepage: https://recognize-anything.github.io/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the Recognize Anything Model (RAM): a strong foundation model for
image tagging. RAM makes a substantial step for large models in computer
vision, demonstrating the zero-shot ability to recognize any common category
with high accuracy. RAM introduces a new paradigm for image tagging, leveraging
large-scale image-text pairs for training instead of manual annotations.
The development of RAM comprises four key steps. Firstly, annotation-free
image tags are obtained at scale through automatic text semantic parsing.
Subsequently, a preliminary model is trained for automatic annotation by
unifying the caption and tagging tasks, supervised by the original texts and
parsed tags, respectively. Thirdly, a data engine is employed to generate
additional annotations and clean incorrect ones. Lastly, the model is retrained
with the processed data and fine-tuned using a smaller but higher-quality
dataset.
We evaluate the tagging capabilities of RAM on numerous benchmarks and
observe impressive zero-shot performance, significantly outperforming CLIP and
BLIP. Remarkably, RAM even surpasses the fully supervised manners and exhibits
competitive performance with the Google tagging API. We are releasing the RAM
at \url{https://recognize-anything.github.io/} to foster the advancements of
large models in computer vision.
| [
{
"version": "v1",
"created": "Tue, 6 Jun 2023 09:00:10 GMT"
},
{
"version": "v2",
"created": "Wed, 7 Jun 2023 04:24:55 GMT"
},
{
"version": "v3",
"created": "Fri, 9 Jun 2023 15:21:06 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Zhang",
"Youcai",
""
],
[
"Huang",
"Xinyu",
""
],
[
"Ma",
"Jinyu",
""
],
[
"Li",
"Zhaoyang",
""
],
[
"Luo",
"Zhaochuan",
""
],
[
"Xie",
"Yanchun",
""
],
[
"Qin",
"Yuzhuo",
""
],
[
"Luo",
"Tong",
""
],
[
"Li",
"Yaqian",
""
],
[
"Liu",
"Shilong",
""
],
[
"Guo",
"Yandong",
""
],
[
"Zhang",
"Lei",
""
]
] | TITLE: Recognize Anything: A Strong Image Tagging Model
ABSTRACT: We present the Recognize Anything Model (RAM): a strong foundation model for
image tagging. RAM makes a substantial step for large models in computer
vision, demonstrating the zero-shot ability to recognize any common category
with high accuracy. RAM introduces a new paradigm for image tagging, leveraging
large-scale image-text pairs for training instead of manual annotations.
The development of RAM comprises four key steps. Firstly, annotation-free
image tags are obtained at scale through automatic text semantic parsing.
Subsequently, a preliminary model is trained for automatic annotation by
unifying the caption and tagging tasks, supervised by the original texts and
parsed tags, respectively. Thirdly, a data engine is employed to generate
additional annotations and clean incorrect ones. Lastly, the model is retrained
with the processed data and fine-tuned using a smaller but higher-quality
dataset.
We evaluate the tagging capabilities of RAM on numerous benchmarks and
observe impressive zero-shot performance, significantly outperforming CLIP and
BLIP. Remarkably, RAM even surpasses the fully supervised manners and exhibits
competitive performance with the Google tagging API. We are releasing the RAM
at \url{https://recognize-anything.github.io/} to foster the advancements of
large models in computer vision.
|
2306.04641 | Jindong Wang | Xin Qin, Jindong Wang, Shuo Ma, Wang Lu, Yongchun Zhu, Xing Xie,
Yiqiang Chen | Generalizable Low-Resource Activity Recognition with Diverse and
Discriminative Representation Learning | Accepted by SIGKDD 2023 Research track; 12 pages; Code is available
at: https://github.com/microsoft/robustlearn | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Human activity recognition (HAR) is a time series classification task that
focuses on identifying the motion patterns from human sensor readings. Adequate
data is essential but a major bottleneck for training a generalizable HAR
model, which assists customization and optimization of online web applications.
However, it is costly in time and economy to collect large-scale labeled data
in reality, i.e., the low-resource challenge. Meanwhile, data collected from
different persons have distribution shifts due to different living habits, body
shapes, age groups, etc. The low-resource and distribution shift challenges are
detrimental to HAR when applying the trained model to new unseen subjects. In
this paper, we propose a novel approach called Diverse and Discriminative
representation Learning (DDLearn) for generalizable low-resource HAR. DDLearn
simultaneously considers diversity and discrimination learning. With the
constructed self-supervised learning task, DDLearn enlarges the data diversity
and explores the latent activity properties. Then, we propose a diversity
preservation module to preserve the diversity of learned features by enlarging
the distribution divergence between the original and augmented domains.
Meanwhile, DDLearn also enhances semantic discrimination by learning
discriminative representations with supervised contrastive learning. Extensive
experiments on three public HAR datasets demonstrate that our method
significantly outperforms state-of-art methods by an average accuracy
improvement of 9.5% under the low-resource distribution shift scenarios, while
being a generic, explainable, and flexible framework. Code is available at:
https://github.com/microsoft/robustlearn.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 08:24:22 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 02:46:34 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Qin",
"Xin",
""
],
[
"Wang",
"Jindong",
""
],
[
"Ma",
"Shuo",
""
],
[
"Lu",
"Wang",
""
],
[
"Zhu",
"Yongchun",
""
],
[
"Xie",
"Xing",
""
],
[
"Chen",
"Yiqiang",
""
]
] | TITLE: Generalizable Low-Resource Activity Recognition with Diverse and
Discriminative Representation Learning
ABSTRACT: Human activity recognition (HAR) is a time series classification task that
focuses on identifying the motion patterns from human sensor readings. Adequate
data is essential but a major bottleneck for training a generalizable HAR
model, which assists customization and optimization of online web applications.
However, it is costly in time and economy to collect large-scale labeled data
in reality, i.e., the low-resource challenge. Meanwhile, data collected from
different persons have distribution shifts due to different living habits, body
shapes, age groups, etc. The low-resource and distribution shift challenges are
detrimental to HAR when applying the trained model to new unseen subjects. In
this paper, we propose a novel approach called Diverse and Discriminative
representation Learning (DDLearn) for generalizable low-resource HAR. DDLearn
simultaneously considers diversity and discrimination learning. With the
constructed self-supervised learning task, DDLearn enlarges the data diversity
and explores the latent activity properties. Then, we propose a diversity
preservation module to preserve the diversity of learned features by enlarging
the distribution divergence between the original and augmented domains.
Meanwhile, DDLearn also enhances semantic discrimination by learning
discriminative representations with supervised contrastive learning. Extensive
experiments on three public HAR datasets demonstrate that our method
significantly outperforms state-of-art methods by an average accuracy
improvement of 9.5% under the low-resource distribution shift scenarios, while
being a generic, explainable, and flexible framework. Code is available at:
https://github.com/microsoft/robustlearn.
|
2306.05434 | Shafiuddin Rehan Ahmed | Shafiuddin Rehan Ahmed, Abhijnan Nath, Michael Regan, Adam Pollins,
Nikhil Krishnaswamy, James H. Martin | How Good is the Model in Model-in-the-loop Event Coreference Resolution
Annotation? | The 17th Liguistics Annotation Workshop, 2023 (LAW-XVII) short paper.
10 pages, 6 figures, 1 table | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Annotating cross-document event coreference links is a time-consuming and
cognitively demanding task that can compromise annotation quality and
efficiency. To address this, we propose a model-in-the-loop annotation approach
for event coreference resolution, where a machine learning model suggests
likely corefering event pairs only. We evaluate the effectiveness of this
approach by first simulating the annotation process and then, using a novel
annotator-centric Recall-Annotation effort trade-off metric, we compare the
results of various underlying models and datasets. We finally present a method
for obtaining 97\% recall while substantially reducing the workload required by
a fully manual annotation process. Code and data can be found at
https://github.com/ahmeshaf/model_in_coref
| [
{
"version": "v1",
"created": "Tue, 6 Jun 2023 18:06:24 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Ahmed",
"Shafiuddin Rehan",
""
],
[
"Nath",
"Abhijnan",
""
],
[
"Regan",
"Michael",
""
],
[
"Pollins",
"Adam",
""
],
[
"Krishnaswamy",
"Nikhil",
""
],
[
"Martin",
"James H.",
""
]
] | TITLE: How Good is the Model in Model-in-the-loop Event Coreference Resolution
Annotation?
ABSTRACT: Annotating cross-document event coreference links is a time-consuming and
cognitively demanding task that can compromise annotation quality and
efficiency. To address this, we propose a model-in-the-loop annotation approach
for event coreference resolution, where a machine learning model suggests
likely corefering event pairs only. We evaluate the effectiveness of this
approach by first simulating the annotation process and then, using a novel
annotator-centric Recall-Annotation effort trade-off metric, we compare the
results of various underlying models and datasets. We finally present a method
for obtaining 97\% recall while substantially reducing the workload required by
a fully manual annotation process. Code and data can be found at
https://github.com/ahmeshaf/model_in_coref
|
2306.05442 | Zhaoyang Huang | Zhaoyang Huang, Xiaoyu Shi, Chao Zhang, Qiang Wang, Yijin Li, Hongwei
Qin, Jifeng Dai, Xiaogang Wang, and Hongsheng Li | FlowFormer: A Transformer Architecture and Its Masked Cost Volume
Autoencoding for Optical Flow | arXiv admin note: substantial text overlap with arXiv:2203.16194,
arXiv:2303.01237 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | This paper introduces a novel transformer-based network architecture,
FlowFormer, along with the Masked Cost Volume AutoEncoding (MCVA) for
pretraining it to tackle the problem of optical flow estimation. FlowFormer
tokenizes the 4D cost-volume built from the source-target image pair and
iteratively refines flow estimation with a cost-volume encoder-decoder
architecture. The cost-volume encoder derives a cost memory with
alternate-group transformer~(AGT) layers in a latent space and the decoder
recurrently decodes flow from the cost memory with dynamic positional cost
queries. On the Sintel benchmark, FlowFormer architecture achieves 1.16 and
2.09 average end-point-error~(AEPE) on the clean and final pass, a 16.5\% and
15.5\% error reduction from the GMA~(1.388 and 2.47). MCVA enhances FlowFormer
by pretraining the cost-volume encoder with a masked autoencoding scheme, which
further unleashes the capability of FlowFormer with unlabeled data. This is
especially critical in optical flow estimation because ground truth flows are
more expensive to acquire than labels in other vision tasks. MCVA improves
FlowFormer all-sided and FlowFormer+MCVA ranks 1st among all published methods
on both Sintel and KITTI-2015 benchmarks and achieves the best generalization
performance. Specifically, FlowFormer+MCVA achieves 1.07 and 1.94 AEPE on the
Sintel benchmark, leading to 7.76\% and 7.18\% error reductions from
FlowFormer.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 12:24:04 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Huang",
"Zhaoyang",
""
],
[
"Shi",
"Xiaoyu",
""
],
[
"Zhang",
"Chao",
""
],
[
"Wang",
"Qiang",
""
],
[
"Li",
"Yijin",
""
],
[
"Qin",
"Hongwei",
""
],
[
"Dai",
"Jifeng",
""
],
[
"Wang",
"Xiaogang",
""
],
[
"Li",
"Hongsheng",
""
]
] | TITLE: FlowFormer: A Transformer Architecture and Its Masked Cost Volume
Autoencoding for Optical Flow
ABSTRACT: This paper introduces a novel transformer-based network architecture,
FlowFormer, along with the Masked Cost Volume AutoEncoding (MCVA) for
pretraining it to tackle the problem of optical flow estimation. FlowFormer
tokenizes the 4D cost-volume built from the source-target image pair and
iteratively refines flow estimation with a cost-volume encoder-decoder
architecture. The cost-volume encoder derives a cost memory with
alternate-group transformer~(AGT) layers in a latent space and the decoder
recurrently decodes flow from the cost memory with dynamic positional cost
queries. On the Sintel benchmark, FlowFormer architecture achieves 1.16 and
2.09 average end-point-error~(AEPE) on the clean and final pass, a 16.5\% and
15.5\% error reduction from the GMA~(1.388 and 2.47). MCVA enhances FlowFormer
by pretraining the cost-volume encoder with a masked autoencoding scheme, which
further unleashes the capability of FlowFormer with unlabeled data. This is
especially critical in optical flow estimation because ground truth flows are
more expensive to acquire than labels in other vision tasks. MCVA improves
FlowFormer all-sided and FlowFormer+MCVA ranks 1st among all published methods
on both Sintel and KITTI-2015 benchmarks and achieves the best generalization
performance. Specifically, FlowFormer+MCVA achieves 1.07 and 1.94 AEPE on the
Sintel benchmark, leading to 7.76\% and 7.18\% error reductions from
FlowFormer.
|
2306.05443 | Qianqian Xie | Qianqian Xie, Weiguang Han, Xiao Zhang, Yanzhao Lai, Min Peng,
Alejandro Lopez-Lira, Jimin Huang | PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark
for Finance | 12 pages, 1 figures | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although large language models (LLMs) has shown great performance on natural
language processing (NLP) in the financial domain, there are no publicly
available financial tailtored LLMs, instruction tuning datasets, and evaluation
benchmarks, which is critical for continually pushing forward the open-source
development of financial artificial intelligence (AI). This paper introduces
PIXIU, a comprehensive framework including the first financial LLM based on
fine-tuning LLaMA with instruction data, the first instruction data with 136K
data samples to support the fine-tuning, and an evaluation benchmark with 5
tasks and 9 datasets. We first construct the large-scale multi-task instruction
data considering a variety of financial tasks, financial document types, and
financial data modalities. We then propose a financial LLM called FinMA by
fine-tuning LLaMA with the constructed dataset to be able to follow
instructions for various financial tasks. To support the evaluation of
financial LLMs, we propose a standardized benchmark that covers a set of
critical financial tasks, including five financial NLP tasks and one financial
prediction task. With this benchmark, we conduct a detailed analysis of FinMA
and several existing LLMs, uncovering their strengths and weaknesses in
handling critical financial tasks. The model, datasets, benchmark, and
experimental results are open-sourced to facilitate future research in
financial AI.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 14:20:29 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Xie",
"Qianqian",
""
],
[
"Han",
"Weiguang",
""
],
[
"Zhang",
"Xiao",
""
],
[
"Lai",
"Yanzhao",
""
],
[
"Peng",
"Min",
""
],
[
"Lopez-Lira",
"Alejandro",
""
],
[
"Huang",
"Jimin",
""
]
] | TITLE: PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark
for Finance
ABSTRACT: Although large language models (LLMs) has shown great performance on natural
language processing (NLP) in the financial domain, there are no publicly
available financial tailtored LLMs, instruction tuning datasets, and evaluation
benchmarks, which is critical for continually pushing forward the open-source
development of financial artificial intelligence (AI). This paper introduces
PIXIU, a comprehensive framework including the first financial LLM based on
fine-tuning LLaMA with instruction data, the first instruction data with 136K
data samples to support the fine-tuning, and an evaluation benchmark with 5
tasks and 9 datasets. We first construct the large-scale multi-task instruction
data considering a variety of financial tasks, financial document types, and
financial data modalities. We then propose a financial LLM called FinMA by
fine-tuning LLaMA with the constructed dataset to be able to follow
instructions for various financial tasks. To support the evaluation of
financial LLMs, we propose a standardized benchmark that covers a set of
critical financial tasks, including five financial NLP tasks and one financial
prediction task. With this benchmark, we conduct a detailed analysis of FinMA
and several existing LLMs, uncovering their strengths and weaknesses in
handling critical financial tasks. The model, datasets, benchmark, and
experimental results are open-sourced to facilitate future research in
financial AI.
|
2306.05446 | Dianna Yee | Colin Lea, Dianna Yee, Jaya Narain, Zifang Huang, Lauren Tooley,
Jeffrey P. Bigham, Leah Findlater | Latent Phrase Matching for Dysarthric Speech | null | null | null | null | eess.AS cs.AI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many consumer speech recognition systems are not tuned for people with speech
disabilities, resulting in poor recognition and user experience, especially for
severe speech differences. Recent studies have emphasized interest in
personalized speech models from people with atypical speech patterns. We
propose a query-by-example-based personalized phrase recognition system that is
trained using small amounts of speech, is language agnostic, does not assume a
traditional pronunciation lexicon, and generalizes well across speech
difference severities. On an internal dataset collected from 32 people with
dysarthria, this approach works regardless of severity and shows a 60%
improvement in recall relative to a commercial speech recognition system. On
the public EasyCall dataset of dysarthric speech, our approach improves
accuracy by 30.5%. Performance degrades as the number of phrases increases, but
consistently outperforms ASR systems when trained with 50 unique phrases.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 17:28:28 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lea",
"Colin",
""
],
[
"Yee",
"Dianna",
""
],
[
"Narain",
"Jaya",
""
],
[
"Huang",
"Zifang",
""
],
[
"Tooley",
"Lauren",
""
],
[
"Bigham",
"Jeffrey P.",
""
],
[
"Findlater",
"Leah",
""
]
] | TITLE: Latent Phrase Matching for Dysarthric Speech
ABSTRACT: Many consumer speech recognition systems are not tuned for people with speech
disabilities, resulting in poor recognition and user experience, especially for
severe speech differences. Recent studies have emphasized interest in
personalized speech models from people with atypical speech patterns. We
propose a query-by-example-based personalized phrase recognition system that is
trained using small amounts of speech, is language agnostic, does not assume a
traditional pronunciation lexicon, and generalizes well across speech
difference severities. On an internal dataset collected from 32 people with
dysarthria, this approach works regardless of severity and shows a 60%
improvement in recall relative to a commercial speech recognition system. On
the public EasyCall dataset of dysarthric speech, our approach improves
accuracy by 30.5%. Performance degrades as the number of phrases increases, but
consistently outperforms ASR systems when trained with 50 unique phrases.
|
2306.05478 | Sajjad Mozaffari | Sajjad Mozaffari, Mreza Alipour Sormoli, Konstantinos Koufos, Graham
Lee, and Mehrdad Dianati | Trajectory Prediction with Observations of Variable-Length for Motion
Planning in Highway Merging scenarios | 8 page, 7 figures, submitted to IEEE ITSC 2023 | null | null | null | cs.RO cs.LG | http://creativecommons.org/licenses/by/4.0/ | Accurate trajectory prediction of nearby vehicles is crucial for the safe
motion planning of automated vehicles in dynamic driving scenarios such as
highway merging. Existing methods cannot initiate prediction for a vehicle
unless observed for a fixed duration of two or more seconds. This prevents a
fast reaction by the ego vehicle to vehicles that enter its perception range,
thus creating safety concerns. Therefore, this paper proposes a novel
transformer-based trajectory prediction approach, specifically trained to
handle any observation length larger than one frame. We perform a comprehensive
evaluation of the proposed method using two large-scale highway trajectory
datasets, namely the highD and exiD. In addition, we study the impact of the
proposed prediction approach on motion planning and control tasks using
extensive merging scenarios from the exiD dataset. To the best of our
knowledge, this marks the first instance where such a large-scale highway
merging dataset has been employed for this purpose. The results demonstrate
that the prediction model achieves state-of-the-art performance on highD
dataset and maintains lower prediction error w.r.t. the constant velocity
across all observation lengths in exiD. Moreover, it significantly enhances
safety, comfort, and efficiency in dense traffic scenarios, as compared to the
constant velocity model.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 18:03:48 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Mozaffari",
"Sajjad",
""
],
[
"Sormoli",
"Mreza Alipour",
""
],
[
"Koufos",
"Konstantinos",
""
],
[
"Lee",
"Graham",
""
],
[
"Dianati",
"Mehrdad",
""
]
] | TITLE: Trajectory Prediction with Observations of Variable-Length for Motion
Planning in Highway Merging scenarios
ABSTRACT: Accurate trajectory prediction of nearby vehicles is crucial for the safe
motion planning of automated vehicles in dynamic driving scenarios such as
highway merging. Existing methods cannot initiate prediction for a vehicle
unless observed for a fixed duration of two or more seconds. This prevents a
fast reaction by the ego vehicle to vehicles that enter its perception range,
thus creating safety concerns. Therefore, this paper proposes a novel
transformer-based trajectory prediction approach, specifically trained to
handle any observation length larger than one frame. We perform a comprehensive
evaluation of the proposed method using two large-scale highway trajectory
datasets, namely the highD and exiD. In addition, we study the impact of the
proposed prediction approach on motion planning and control tasks using
extensive merging scenarios from the exiD dataset. To the best of our
knowledge, this marks the first instance where such a large-scale highway
merging dataset has been employed for this purpose. The results demonstrate
that the prediction model achieves state-of-the-art performance on highD
dataset and maintains lower prediction error w.r.t. the constant velocity
across all observation lengths in exiD. Moreover, it significantly enhances
safety, comfort, and efficiency in dense traffic scenarios, as compared to the
constant velocity model.
|
2306.05483 | YiDing Jiang | Yiding Jiang, J. Zico Kolter, Roberta Raileanu | On the Importance of Exploration for Generalization in Reinforcement
Learning | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Existing approaches for improving generalization in deep reinforcement
learning (RL) have mostly focused on representation learning, neglecting
RL-specific aspects such as exploration. We hypothesize that the agent's
exploration strategy plays a key role in its ability to generalize to new
environments. Through a series of experiments in a tabular contextual MDP, we
show that exploration is helpful not only for efficiently finding the optimal
policy for the training environments but also for acquiring knowledge that
helps decision making in unseen environments. Based on these observations, we
propose EDE: Exploration via Distributional Ensemble, a method that encourages
exploration of states with high epistemic uncertainty through an ensemble of
Q-value distributions. Our algorithm is the first value-based approach to
achieve state-of-the-art on both Procgen and Crafter, two benchmarks for
generalization in RL with high-dimensional observations. The open-sourced
implementation can be found at https://github.com/facebookresearch/ede .
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 18:07:02 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Jiang",
"Yiding",
""
],
[
"Kolter",
"J. Zico",
""
],
[
"Raileanu",
"Roberta",
""
]
] | TITLE: On the Importance of Exploration for Generalization in Reinforcement
Learning
ABSTRACT: Existing approaches for improving generalization in deep reinforcement
learning (RL) have mostly focused on representation learning, neglecting
RL-specific aspects such as exploration. We hypothesize that the agent's
exploration strategy plays a key role in its ability to generalize to new
environments. Through a series of experiments in a tabular contextual MDP, we
show that exploration is helpful not only for efficiently finding the optimal
policy for the training environments but also for acquiring knowledge that
helps decision making in unseen environments. Based on these observations, we
propose EDE: Exploration via Distributional Ensemble, a method that encourages
exploration of states with high epistemic uncertainty through an ensemble of
Q-value distributions. Our algorithm is the first value-based approach to
achieve state-of-the-art on both Procgen and Crafter, two benchmarks for
generalization in RL with high-dimensional observations. The open-sourced
implementation can be found at https://github.com/facebookresearch/ede .
|
2306.05484 | Christopher Frye | Bethany Connolly, Kim Moore, Tobias Schwedes, Alexander Adam, Gary
Willis, Ilya Feige, Christopher Frye | Task-specific experimental design for treatment effect estimation | To appear in ICML 2023; 8 pages, 7 figures, 4 appendices | null | null | null | stat.ME cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding causality should be a core requirement of any attempt to build
real impact through AI. Due to the inherent unobservability of counterfactuals,
large randomised trials (RCTs) are the standard for causal inference. But large
experiments are generically expensive, and randomisation carries its own costs,
e.g. when suboptimal decisions are trialed. Recent work has proposed more
sample-efficient alternatives to RCTs, but these are not adaptable to the
downstream application for which the causal effect is sought. In this work, we
develop a task-specific approach to experimental design and derive sampling
strategies customised to particular downstream applications. Across a range of
important tasks, real-world datasets, and sample sizes, our method outperforms
other benchmarks, e.g. requiring an order-of-magnitude less data to match RCT
performance on targeted marketing tasks.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 18:10:37 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Connolly",
"Bethany",
""
],
[
"Moore",
"Kim",
""
],
[
"Schwedes",
"Tobias",
""
],
[
"Adam",
"Alexander",
""
],
[
"Willis",
"Gary",
""
],
[
"Feige",
"Ilya",
""
],
[
"Frye",
"Christopher",
""
]
] | TITLE: Task-specific experimental design for treatment effect estimation
ABSTRACT: Understanding causality should be a core requirement of any attempt to build
real impact through AI. Due to the inherent unobservability of counterfactuals,
large randomised trials (RCTs) are the standard for causal inference. But large
experiments are generically expensive, and randomisation carries its own costs,
e.g. when suboptimal decisions are trialed. Recent work has proposed more
sample-efficient alternatives to RCTs, but these are not adaptable to the
downstream application for which the causal effect is sought. In this work, we
develop a task-specific approach to experimental design and derive sampling
strategies customised to particular downstream applications. Across a range of
important tasks, real-world datasets, and sample sizes, our method outperforms
other benchmarks, e.g. requiring an order-of-magnitude less data to match RCT
performance on targeted marketing tasks.
|
2306.05493 | Prannay Kaul | Prannay Kaul, Weidi Xie, Andrew Zisserman | Multi-Modal Classifiers for Open-Vocabulary Object Detection | ICML 2023, project page:
https://www.robots.ox.ac.uk/vgg/research/mm-ovod/ | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The goal of this paper is open-vocabulary object detection (OVOD)
$\unicode{x2013}$ building a model that can detect objects beyond the set of
categories seen at training, thus enabling the user to specify categories of
interest at inference without the need for model retraining. We adopt a
standard two-stage object detector architecture, and explore three ways for
specifying novel categories: via language descriptions, via image exemplars, or
via a combination of the two. We make three contributions: first, we prompt a
large language model (LLM) to generate informative language descriptions for
object classes, and construct powerful text-based classifiers; second, we
employ a visual aggregator on image exemplars that can ingest any number of
images as input, forming vision-based classifiers; and third, we provide a
simple method to fuse information from language descriptions and image
exemplars, yielding a multi-modal classifier. When evaluating on the
challenging LVIS open-vocabulary benchmark we demonstrate that: (i) our
text-based classifiers outperform all previous OVOD works; (ii) our
vision-based classifiers perform as well as text-based classifiers in prior
work; (iii) using multi-modal classifiers perform better than either modality
alone; and finally, (iv) our text-based and multi-modal classifiers yield
better performance than a fully-supervised detector.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 18:31:56 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Kaul",
"Prannay",
""
],
[
"Xie",
"Weidi",
""
],
[
"Zisserman",
"Andrew",
""
]
] | TITLE: Multi-Modal Classifiers for Open-Vocabulary Object Detection
ABSTRACT: The goal of this paper is open-vocabulary object detection (OVOD)
$\unicode{x2013}$ building a model that can detect objects beyond the set of
categories seen at training, thus enabling the user to specify categories of
interest at inference without the need for model retraining. We adopt a
standard two-stage object detector architecture, and explore three ways for
specifying novel categories: via language descriptions, via image exemplars, or
via a combination of the two. We make three contributions: first, we prompt a
large language model (LLM) to generate informative language descriptions for
object classes, and construct powerful text-based classifiers; second, we
employ a visual aggregator on image exemplars that can ingest any number of
images as input, forming vision-based classifiers; and third, we provide a
simple method to fuse information from language descriptions and image
exemplars, yielding a multi-modal classifier. When evaluating on the
challenging LVIS open-vocabulary benchmark we demonstrate that: (i) our
text-based classifiers outperform all previous OVOD works; (ii) our
vision-based classifiers perform as well as text-based classifiers in prior
work; (iii) using multi-modal classifiers perform better than either modality
alone; and finally, (iv) our text-based and multi-modal classifiers yield
better performance than a fully-supervised detector.
|
2306.05495 | Abhishri Ajit Medewar | Abhishri Ajit Medewar, Swanand Ashokrao Kavitkar | Is Attentional Channel Processing Design Required? Comprehensive
Analysis Of Robustness Between Vision Transformers And Fully Attentional
Networks | 4 pages, 12 figures | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | The robustness testing has been performed for standard CNN models and Vision
Transformers, however there is a lack of comprehensive study between the
robustness of traditional Vision Transformers without an extra attentional
channel design and the latest fully attentional network(FAN) models. So in this
paper, we use the ImageNet dataset to compare the robustness of fully
attentional network(FAN) models with traditional Vision Transformers to
understand the role of an attentional channel processing design using white box
attacks and also study the transferability between the same using black box
attacks.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 18:33:12 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Medewar",
"Abhishri Ajit",
""
],
[
"Kavitkar",
"Swanand Ashokrao",
""
]
] | TITLE: Is Attentional Channel Processing Design Required? Comprehensive
Analysis Of Robustness Between Vision Transformers And Fully Attentional
Networks
ABSTRACT: The robustness testing has been performed for standard CNN models and Vision
Transformers, however there is a lack of comprehensive study between the
robustness of traditional Vision Transformers without an extra attentional
channel design and the latest fully attentional network(FAN) models. So in this
paper, we use the ImageNet dataset to compare the robustness of fully
attentional network(FAN) models with traditional Vision Transformers to
understand the role of an attentional channel processing design using white box
attacks and also study the transferability between the same using black box
attacks.
|
2306.05500 | Lucas Monteiro Paes | Alexander Lin, Lucas Monteiro Paes, Sree Harsha Tanneru, Suraj
Srinivas, Himabindu Lakkaraju | Word-Level Explanations for Analyzing Bias in Text-to-Image Models | 5 main pages, 3 pages in appendix, and 3 figures | null | null | null | cs.CL cs.AI cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Text-to-image models take a sentence (i.e., prompt) and generate images
associated with this input prompt. These models have created award wining-art,
videos, and even synthetic datasets. However, text-to-image (T2I) models can
generate images that underrepresent minorities based on race and sex. This
paper investigates which word in the input prompt is responsible for bias in
generated images. We introduce a method for computing scores for each word in
the prompt; these scores represent its influence on biases in the model's
output. Our method follows the principle of \emph{explaining by removing},
leveraging masked language models to calculate the influence scores. We perform
experiments on Stable Diffusion to demonstrate that our method identifies the
replication of societal stereotypes in generated images.
| [
{
"version": "v1",
"created": "Sat, 3 Jun 2023 21:39:07 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lin",
"Alexander",
""
],
[
"Paes",
"Lucas Monteiro",
""
],
[
"Tanneru",
"Sree Harsha",
""
],
[
"Srinivas",
"Suraj",
""
],
[
"Lakkaraju",
"Himabindu",
""
]
] | TITLE: Word-Level Explanations for Analyzing Bias in Text-to-Image Models
ABSTRACT: Text-to-image models take a sentence (i.e., prompt) and generate images
associated with this input prompt. These models have created award wining-art,
videos, and even synthetic datasets. However, text-to-image (T2I) models can
generate images that underrepresent minorities based on race and sex. This
paper investigates which word in the input prompt is responsible for bias in
generated images. We introduce a method for computing scores for each word in
the prompt; these scores represent its influence on biases in the model's
output. Our method follows the principle of \emph{explaining by removing},
leveraging masked language models to calculate the influence scores. We perform
experiments on Stable Diffusion to demonstrate that our method identifies the
replication of societal stereotypes in generated images.
|
2306.05514 | Mansoor Ahmed | Mansoor Ahmed, Usama Sardar, Sarwan Ali, Shafiq Alam, Murray
Patterson, Imdad Ullah Khan | Robust Brain Age Estimation via Regression Models and MRI-derived
Features | Published at the 15th International Conference on Computational
Collective Intelligence | null | null | null | eess.IV cs.CV cs.LG q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The determination of biological brain age is a crucial biomarker in the
assessment of neurological disorders and understanding of the morphological
changes that occur during aging. Various machine learning models have been
proposed for estimating brain age through Magnetic Resonance Imaging (MRI) of
healthy controls. However, developing a robust brain age estimation (BAE)
framework has been challenging due to the selection of appropriate MRI-derived
features and the high cost of MRI acquisition. In this study, we present a
novel BAE framework using the Open Big Healthy Brain (OpenBHB) dataset, which
is a new multi-site and publicly available benchmark dataset that includes
region-wise feature metrics derived from T1-weighted (T1-w) brain MRI scans of
3965 healthy controls aged between 6 to 86 years. Our approach integrates three
different MRI-derived region-wise features and different regression models,
resulting in a highly accurate brain age estimation with a Mean Absolute Error
(MAE) of 3.25 years, demonstrating the framework's robustness. We also analyze
our model's regression-based performance on gender-wise (male and female)
healthy test groups. The proposed BAE framework provides a new approach for
estimating brain age, which has important implications for the understanding of
neurological disorders and age-related brain changes.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 19:07:22 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Ahmed",
"Mansoor",
""
],
[
"Sardar",
"Usama",
""
],
[
"Ali",
"Sarwan",
""
],
[
"Alam",
"Shafiq",
""
],
[
"Patterson",
"Murray",
""
],
[
"Khan",
"Imdad Ullah",
""
]
] | TITLE: Robust Brain Age Estimation via Regression Models and MRI-derived
Features
ABSTRACT: The determination of biological brain age is a crucial biomarker in the
assessment of neurological disorders and understanding of the morphological
changes that occur during aging. Various machine learning models have been
proposed for estimating brain age through Magnetic Resonance Imaging (MRI) of
healthy controls. However, developing a robust brain age estimation (BAE)
framework has been challenging due to the selection of appropriate MRI-derived
features and the high cost of MRI acquisition. In this study, we present a
novel BAE framework using the Open Big Healthy Brain (OpenBHB) dataset, which
is a new multi-site and publicly available benchmark dataset that includes
region-wise feature metrics derived from T1-weighted (T1-w) brain MRI scans of
3965 healthy controls aged between 6 to 86 years. Our approach integrates three
different MRI-derived region-wise features and different regression models,
resulting in a highly accurate brain age estimation with a Mean Absolute Error
(MAE) of 3.25 years, demonstrating the framework's robustness. We also analyze
our model's regression-based performance on gender-wise (male and female)
healthy test groups. The proposed BAE framework provides a new approach for
estimating brain age, which has important implications for the understanding of
neurological disorders and age-related brain changes.
|
2306.05528 | Marc Huertas-Company | Marc Huertas-Company, Regina Sarmiento, Johan Knapen | A brief review of contrastive learning applied to astrophysics | Invited review to be published in RASTI | null | null | null | astro-ph.IM cs.LG | http://creativecommons.org/licenses/by/4.0/ | Reliable tools to extract patterns from high-dimensionality spaces are
becoming more necessary as astronomical datasets increase both in volume and
complexity. Contrastive Learning is a self-supervised machine learning
algorithm that extracts informative measurements from multi-dimensional
datasets, which has become increasingly popular in the computer vision and
Machine Learning communities in recent years. To do so, it maximizes the
agreement between the information extracted from augmented versions of the same
input data, making the final representation invariant to the applied
transformations. Contrastive Learning is particularly useful in astronomy for
removing known instrumental effects and for performing supervised
classifications and regressions with a limited amount of available labels,
showing a promising avenue towards \emph{Foundation Models}. This short review
paper briefly summarizes the main concepts behind contrastive learning and
reviews the first promising applications to astronomy. We include some
practical recommendations on which applications are particularly attractive for
contrastive learning.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 19:56:32 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Huertas-Company",
"Marc",
""
],
[
"Sarmiento",
"Regina",
""
],
[
"Knapen",
"Johan",
""
]
] | TITLE: A brief review of contrastive learning applied to astrophysics
ABSTRACT: Reliable tools to extract patterns from high-dimensionality spaces are
becoming more necessary as astronomical datasets increase both in volume and
complexity. Contrastive Learning is a self-supervised machine learning
algorithm that extracts informative measurements from multi-dimensional
datasets, which has become increasingly popular in the computer vision and
Machine Learning communities in recent years. To do so, it maximizes the
agreement between the information extracted from augmented versions of the same
input data, making the final representation invariant to the applied
transformations. Contrastive Learning is particularly useful in astronomy for
removing known instrumental effects and for performing supervised
classifications and regressions with a limited amount of available labels,
showing a promising avenue towards \emph{Foundation Models}. This short review
paper briefly summarizes the main concepts behind contrastive learning and
reviews the first promising applications to astronomy. We include some
practical recommendations on which applications are particularly attractive for
contrastive learning.
|
2306.05537 | Guan Wang | Guan Wang, Weihua Li, Edmund M-K. Lai, Quan Bai | AaKOS: Aspect-adaptive Knowledge-based Opinion Summarization | 21 pages, 4 figures, 7 tables | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | The rapid growth of information on the Internet has led to an overwhelming
amount of opinions and comments on various activities, products, and services.
This makes it difficult and time-consuming for users to process all the
available information when making decisions. Text summarization, a Natural
Language Processing (NLP) task, has been widely explored to help users quickly
retrieve relevant information by generating short and salient content from long
or multiple documents. Recent advances in pre-trained language models, such as
ChatGPT, have demonstrated the potential of Large Language Models (LLMs) in
text generation. However, LLMs require massive amounts of data and resources
and are challenging to implement as offline applications. Furthermore, existing
text summarization approaches often lack the ``adaptive" nature required to
capture diverse aspects in opinion summarization, which is particularly
detrimental to users with specific requirements or preferences. In this paper,
we propose an Aspect-adaptive Knowledge-based Opinion Summarization model for
product reviews, which effectively captures the adaptive nature required for
opinion summarization. The model generates aspect-oriented summaries given a
set of reviews for a particular product, efficiently providing users with
useful information on specific aspects they are interested in, ensuring the
generated summaries are more personalized and informative. Extensive
experiments have been conducted using real-world datasets to evaluate the
proposed model. The results demonstrate that our model outperforms
state-of-the-art approaches and is adaptive and efficient in generating
summaries that focus on particular aspects, enabling users to make
well-informed decisions and catering to their diverse interests and
preferences.
| [
{
"version": "v1",
"created": "Fri, 26 May 2023 03:44:35 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wang",
"Guan",
""
],
[
"Li",
"Weihua",
""
],
[
"Lai",
"Edmund M-K.",
""
],
[
"Bai",
"Quan",
""
]
] | TITLE: AaKOS: Aspect-adaptive Knowledge-based Opinion Summarization
ABSTRACT: The rapid growth of information on the Internet has led to an overwhelming
amount of opinions and comments on various activities, products, and services.
This makes it difficult and time-consuming for users to process all the
available information when making decisions. Text summarization, a Natural
Language Processing (NLP) task, has been widely explored to help users quickly
retrieve relevant information by generating short and salient content from long
or multiple documents. Recent advances in pre-trained language models, such as
ChatGPT, have demonstrated the potential of Large Language Models (LLMs) in
text generation. However, LLMs require massive amounts of data and resources
and are challenging to implement as offline applications. Furthermore, existing
text summarization approaches often lack the ``adaptive" nature required to
capture diverse aspects in opinion summarization, which is particularly
detrimental to users with specific requirements or preferences. In this paper,
we propose an Aspect-adaptive Knowledge-based Opinion Summarization model for
product reviews, which effectively captures the adaptive nature required for
opinion summarization. The model generates aspect-oriented summaries given a
set of reviews for a particular product, efficiently providing users with
useful information on specific aspects they are interested in, ensuring the
generated summaries are more personalized and informative. Extensive
experiments have been conducted using real-world datasets to evaluate the
proposed model. The results demonstrate that our model outperforms
state-of-the-art approaches and is adaptive and efficient in generating
summaries that focus on particular aspects, enabling users to make
well-informed decisions and catering to their diverse interests and
preferences.
|
2306.05540 | Preslav Nakov | Jinyan Su, Terry Yue Zhuo, Di Wang, Preslav Nakov | DetectLLM: Leveraging Log Rank Information for Zero-Shot Detection of
Machine-Generated Text | machine-generated text, large language models, LLMs, zero-shot | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the rapid progress of large language models (LLMs) and the huge amount
of text they generated, it becomes more and more impractical to manually
distinguish whether a text is machine-generated. Given the growing use of LLMs
in social media and education, it prompts us to develop methods to detect
machine-generated text, preventing malicious usage such as plagiarism,
misinformation, and propaganda. Previous work has studied several zero-shot
methods, which require no training data. These methods achieve good
performance, but there is still a lot of room for improvement. In this paper,
we introduce two novel zero-shot methods for detecting machine-generated text
by leveraging the log rank information. One is called DetectLLM-LRR, which is
fast and efficient, and the other is called DetectLLM-NPR, which is more
accurate, but slower due to the need for perturbations. Our experiments on
three datasets and seven language models show that our proposed methods improve
over the state of the art by 3.9 and 1.75 AUROC points absolute. Moreover,
DetectLLM-NPR needs fewer perturbations than previous work to achieve the same
level of performance, which makes it more practical for real-world use. We also
investigate the efficiency--performance trade-off based on users preference on
these two measures and we provide intuition for using them in practice
effectively. We release the data and the code of both methods in
https://github.com/mbzuai-nlp/DetectLLM
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 11:18:30 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Su",
"Jinyan",
""
],
[
"Zhuo",
"Terry Yue",
""
],
[
"Wang",
"Di",
""
],
[
"Nakov",
"Preslav",
""
]
] | TITLE: DetectLLM: Leveraging Log Rank Information for Zero-Shot Detection of
Machine-Generated Text
ABSTRACT: With the rapid progress of large language models (LLMs) and the huge amount
of text they generated, it becomes more and more impractical to manually
distinguish whether a text is machine-generated. Given the growing use of LLMs
in social media and education, it prompts us to develop methods to detect
machine-generated text, preventing malicious usage such as plagiarism,
misinformation, and propaganda. Previous work has studied several zero-shot
methods, which require no training data. These methods achieve good
performance, but there is still a lot of room for improvement. In this paper,
we introduce two novel zero-shot methods for detecting machine-generated text
by leveraging the log rank information. One is called DetectLLM-LRR, which is
fast and efficient, and the other is called DetectLLM-NPR, which is more
accurate, but slower due to the need for perturbations. Our experiments on
three datasets and seven language models show that our proposed methods improve
over the state of the art by 3.9 and 1.75 AUROC points absolute. Moreover,
DetectLLM-NPR needs fewer perturbations than previous work to achieve the same
level of performance, which makes it more practical for real-world use. We also
investigate the efficiency--performance trade-off based on users preference on
these two measures and we provide intuition for using them in practice
effectively. We release the data and the code of both methods in
https://github.com/mbzuai-nlp/DetectLLM
|
2306.05544 | Lingjie Liu | Jiatao Gu, Shuangfei Zhai, Yizhe Zhang, Lingjie Liu, Josh Susskind | BOOT: Data-free Distillation of Denoising Diffusion Models with
Bootstrapping | In progress | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion models have demonstrated excellent potential for generating diverse
images. However, their performance often suffers from slow generation due to
iterative denoising. Knowledge distillation has been recently proposed as a
remedy that can reduce the number of inference steps to one or a few without
significant quality degradation. However, existing distillation methods either
require significant amounts of offline computation for generating synthetic
training data from the teacher model or need to perform expensive online
learning with the help of real data. In this work, we present a novel technique
called BOOT, that overcomes these limitations with an efficient data-free
distillation algorithm. The core idea is to learn a time-conditioned model that
predicts the output of a pre-trained diffusion model teacher given any time
step. Such a model can be efficiently trained based on bootstrapping from two
consecutive sampled steps. Furthermore, our method can be easily adapted to
large-scale text-to-image diffusion models, which are challenging for
conventional methods given the fact that the training sets are often large and
difficult to access. We demonstrate the effectiveness of our approach on
several benchmark datasets in the DDIM setting, achieving comparable generation
quality while being orders of magnitude faster than the diffusion teacher. The
text-to-image results show that the proposed approach is able to handle highly
complex distributions, shedding light on more efficient generative modeling.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 20:30:55 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Gu",
"Jiatao",
""
],
[
"Zhai",
"Shuangfei",
""
],
[
"Zhang",
"Yizhe",
""
],
[
"Liu",
"Lingjie",
""
],
[
"Susskind",
"Josh",
""
]
] | TITLE: BOOT: Data-free Distillation of Denoising Diffusion Models with
Bootstrapping
ABSTRACT: Diffusion models have demonstrated excellent potential for generating diverse
images. However, their performance often suffers from slow generation due to
iterative denoising. Knowledge distillation has been recently proposed as a
remedy that can reduce the number of inference steps to one or a few without
significant quality degradation. However, existing distillation methods either
require significant amounts of offline computation for generating synthetic
training data from the teacher model or need to perform expensive online
learning with the help of real data. In this work, we present a novel technique
called BOOT, that overcomes these limitations with an efficient data-free
distillation algorithm. The core idea is to learn a time-conditioned model that
predicts the output of a pre-trained diffusion model teacher given any time
step. Such a model can be efficiently trained based on bootstrapping from two
consecutive sampled steps. Furthermore, our method can be easily adapted to
large-scale text-to-image diffusion models, which are challenging for
conventional methods given the fact that the training sets are often large and
difficult to access. We demonstrate the effectiveness of our approach on
several benchmark datasets in the DDIM setting, achieving comparable generation
quality while being orders of magnitude faster than the diffusion teacher. The
text-to-image results show that the proposed approach is able to handle highly
complex distributions, shedding light on more efficient generative modeling.
|
2306.05556 | Justin Xie | Justin J. Xie and Ameeta Agrawal | Emotion and Sentiment Guided Paraphrasing | 13th Workshop on Computational Approaches to Subjectivity, Sentiment
& Social Media Analysis (WASSA) 2023 at The 61st Annual Meeting of the
Association for Computational Linguistics (ACL) 2023. arXiv admin note:
substantial text overlap with arXiv:2212.03297 | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Paraphrase generation, a.k.a. paraphrasing, is a common and important task in
natural language processing. Emotional paraphrasing, which changes the emotion
embodied in a piece of text while preserving its meaning, has many potential
applications, including moderating online dialogues and preventing
cyberbullying. We introduce a new task of fine-grained emotional paraphrasing
along emotion gradients, that is, altering the emotional intensities of the
paraphrases in fine-grained settings following smooth variations in affective
dimensions while preserving the meaning of the original text. We reconstruct
several widely used paraphrasing datasets by augmenting the input and target
texts with their fine-grained emotion labels. Then, we propose a framework for
emotion and sentiment guided paraphrasing by leveraging pre-trained language
models for conditioned text generation. Extensive evaluation of the fine-tuned
models suggests that including fine-grained emotion labels in the paraphrase
task significantly improves the likelihood of obtaining high-quality
paraphrases that reflect the desired emotions while achieving consistently
better scores in paraphrase metrics such as BLEU, ROUGE, and METEOR.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 20:59:40 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Xie",
"Justin J.",
""
],
[
"Agrawal",
"Ameeta",
""
]
] | TITLE: Emotion and Sentiment Guided Paraphrasing
ABSTRACT: Paraphrase generation, a.k.a. paraphrasing, is a common and important task in
natural language processing. Emotional paraphrasing, which changes the emotion
embodied in a piece of text while preserving its meaning, has many potential
applications, including moderating online dialogues and preventing
cyberbullying. We introduce a new task of fine-grained emotional paraphrasing
along emotion gradients, that is, altering the emotional intensities of the
paraphrases in fine-grained settings following smooth variations in affective
dimensions while preserving the meaning of the original text. We reconstruct
several widely used paraphrasing datasets by augmenting the input and target
texts with their fine-grained emotion labels. Then, we propose a framework for
emotion and sentiment guided paraphrasing by leveraging pre-trained language
models for conditioned text generation. Extensive evaluation of the fine-tuned
models suggests that including fine-grained emotion labels in the paraphrase
task significantly improves the likelihood of obtaining high-quality
paraphrases that reflect the desired emotions while achieving consistently
better scores in paraphrase metrics such as BLEU, ROUGE, and METEOR.
|
2306.05561 | Oleksandr Yermilov | Oleksandr Yermilov, Vipul Raheja, Artem Chernodub | Privacy- and Utility-Preserving NLP with Anonymized Data: A case study
of Pseudonymization | 10 pages. Accepted for TrustNLP workshop at ACL2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This work investigates the effectiveness of different pseudonymization
techniques, ranging from rule-based substitutions to using pre-trained Large
Language Models (LLMs), on a variety of datasets and models used for two widely
used NLP tasks: text classification and summarization. Our work provides
crucial insights into the gaps between original and anonymized data (focusing
on the pseudonymization technique) and model quality and fosters future
research into higher-quality anonymization techniques to better balance the
trade-offs between data protection and utility preservation. We make our code,
pseudonymized datasets, and downstream models publicly available
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 21:06:19 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Yermilov",
"Oleksandr",
""
],
[
"Raheja",
"Vipul",
""
],
[
"Chernodub",
"Artem",
""
]
] | TITLE: Privacy- and Utility-Preserving NLP with Anonymized Data: A case study
of Pseudonymization
ABSTRACT: This work investigates the effectiveness of different pseudonymization
techniques, ranging from rule-based substitutions to using pre-trained Large
Language Models (LLMs), on a variety of datasets and models used for two widely
used NLP tasks: text classification and summarization. Our work provides
crucial insights into the gaps between original and anonymized data (focusing
on the pseudonymization technique) and model quality and fosters future
research into higher-quality anonymization techniques to better balance the
trade-offs between data protection and utility preservation. We make our code,
pseudonymized datasets, and downstream models publicly available
|
2306.05562 | Adam Cobb | Adam D. Cobb, Anirban Roy, Daniel Elenius, F. Michael Heim, Brian
Swenson, Sydney Whittington, James D. Walker, Theodore Bapty, Joseph Hite,
Karthik Ramani, Christopher McComb, Susmit Jha | AircraftVerse: A Large-Scale Multimodal Dataset of Aerial Vehicle
Designs | The dataset is hosted at https://zenodo.org/record/6525446, baseline
models and code at https://github.com/SRI-CSL/AircraftVerse, and the dataset
description at https://aircraftverse.onrender.com/ | null | null | null | cs.RO cs.AI cs.CE | http://creativecommons.org/licenses/by-sa/4.0/ | We present AircraftVerse, a publicly available aerial vehicle design dataset.
Aircraft design encompasses different physics domains and, hence, multiple
modalities of representation. The evaluation of these cyber-physical system
(CPS) designs requires the use of scientific analytical and simulation models
ranging from computer-aided design tools for structural and manufacturing
analysis, computational fluid dynamics tools for drag and lift computation,
battery models for energy estimation, and simulation models for flight control
and dynamics. AircraftVerse contains 27,714 diverse air vehicle designs - the
largest corpus of engineering designs with this level of complexity. Each
design comprises the following artifacts: a symbolic design tree describing
topology, propulsion subsystem, battery subsystem, and other design details; a
STandard for the Exchange of Product (STEP) model data; a 3D CAD design using a
stereolithography (STL) file format; a 3D point cloud for the shape of the
design; and evaluation results from high fidelity state-of-the-art physics
models that characterize performance metrics such as maximum flight distance
and hover-time. We also present baseline surrogate models that use different
modalities of design representation to predict design performance metrics,
which we provide as part of our dataset release. Finally, we discuss the
potential impact of this dataset on the use of learning in aircraft design and,
more generally, in CPS. AircraftVerse is accompanied by a data card, and it is
released under Creative Commons Attribution-ShareAlike (CC BY-SA) license. The
dataset is hosted at https://zenodo.org/record/6525446, baseline models and
code at https://github.com/SRI-CSL/AircraftVerse, and the dataset description
at https://aircraftverse.onrender.com/.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 21:07:15 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Cobb",
"Adam D.",
""
],
[
"Roy",
"Anirban",
""
],
[
"Elenius",
"Daniel",
""
],
[
"Heim",
"F. Michael",
""
],
[
"Swenson",
"Brian",
""
],
[
"Whittington",
"Sydney",
""
],
[
"Walker",
"James D.",
""
],
[
"Bapty",
"Theodore",
""
],
[
"Hite",
"Joseph",
""
],
[
"Ramani",
"Karthik",
""
],
[
"McComb",
"Christopher",
""
],
[
"Jha",
"Susmit",
""
]
] | TITLE: AircraftVerse: A Large-Scale Multimodal Dataset of Aerial Vehicle
Designs
ABSTRACT: We present AircraftVerse, a publicly available aerial vehicle design dataset.
Aircraft design encompasses different physics domains and, hence, multiple
modalities of representation. The evaluation of these cyber-physical system
(CPS) designs requires the use of scientific analytical and simulation models
ranging from computer-aided design tools for structural and manufacturing
analysis, computational fluid dynamics tools for drag and lift computation,
battery models for energy estimation, and simulation models for flight control
and dynamics. AircraftVerse contains 27,714 diverse air vehicle designs - the
largest corpus of engineering designs with this level of complexity. Each
design comprises the following artifacts: a symbolic design tree describing
topology, propulsion subsystem, battery subsystem, and other design details; a
STandard for the Exchange of Product (STEP) model data; a 3D CAD design using a
stereolithography (STL) file format; a 3D point cloud for the shape of the
design; and evaluation results from high fidelity state-of-the-art physics
models that characterize performance metrics such as maximum flight distance
and hover-time. We also present baseline surrogate models that use different
modalities of design representation to predict design performance metrics,
which we provide as part of our dataset release. Finally, we discuss the
potential impact of this dataset on the use of learning in aircraft design and,
more generally, in CPS. AircraftVerse is accompanied by a data card, and it is
released under Creative Commons Attribution-ShareAlike (CC BY-SA) license. The
dataset is hosted at https://zenodo.org/record/6525446, baseline models and
code at https://github.com/SRI-CSL/AircraftVerse, and the dataset description
at https://aircraftverse.onrender.com/.
|
2306.05592 | Baihe Huang | Baihe Huang, Sai Praneeth Karimireddy, Michael I. Jordan | Evaluating and Incentivizing Diverse Data Contributions in Collaborative
Learning | null | null | null | null | cs.GT cs.CY cs.DC cs.LG econ.TH | http://creativecommons.org/publicdomain/zero/1.0/ | For a federated learning model to perform well, it is crucial to have a
diverse and representative dataset. However, the data contributors may only be
concerned with the performance on a specific subset of the population, which
may not reflect the diversity of the wider population. This creates a tension
between the principal (the FL platform designer) who cares about global
performance and the agents (the data collectors) who care about local
performance. In this work, we formulate this tension as a game between the
principal and multiple agents, and focus on the linear experiment design
problem to formally study their interaction. We show that the statistical
criterion used to quantify the diversity of the data, as well as the choice of
the federated learning algorithm used, has a significant effect on the
resulting equilibrium. We leverage this to design simple optimal federated
learning mechanisms that encourage data collectors to contribute data
representative of the global population, thereby maximizing global performance.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 23:38:25 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Huang",
"Baihe",
""
],
[
"Karimireddy",
"Sai Praneeth",
""
],
[
"Jordan",
"Michael I.",
""
]
] | TITLE: Evaluating and Incentivizing Diverse Data Contributions in Collaborative
Learning
ABSTRACT: For a federated learning model to perform well, it is crucial to have a
diverse and representative dataset. However, the data contributors may only be
concerned with the performance on a specific subset of the population, which
may not reflect the diversity of the wider population. This creates a tension
between the principal (the FL platform designer) who cares about global
performance and the agents (the data collectors) who care about local
performance. In this work, we formulate this tension as a game between the
principal and multiple agents, and focus on the linear experiment design
problem to formally study their interaction. We show that the statistical
criterion used to quantify the diversity of the data, as well as the choice of
the federated learning algorithm used, has a significant effect on the
resulting equilibrium. We leverage this to design simple optimal federated
learning mechanisms that encourage data collectors to contribute data
representative of the global population, thereby maximizing global performance.
|
2306.05596 | Muskan Garg | Muskan Garg, Manas Gaur, Raxit Goswami, Sunghwan Sohn | LOST: A Mental Health Dataset of Low Self-esteem in Reddit Posts | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Low self-esteem and interpersonal needs (i.e., thwarted belongingness (TB)
and perceived burdensomeness (PB)) have a major impact on depression and
suicide attempts. Individuals seek social connectedness on social media to
boost and alleviate their loneliness. Social media platforms allow people to
express their thoughts, experiences, beliefs, and emotions. Prior studies on
mental health from social media have focused on symptoms, causes, and
disorders. Whereas an initial screening of social media content for
interpersonal risk factors and low self-esteem may raise early alerts and
assign therapists to at-risk users of mental disturbance. Standardized scales
measure self-esteem and interpersonal needs from questions created using
psychological theories. In the current research, we introduce a
psychology-grounded and expertly annotated dataset, LoST: Low Self esTeem, to
study and detect low self-esteem on Reddit. Through an annotation approach
involving checks on coherence, correctness, consistency, and reliability, we
ensure gold-standard for supervised learning. We present results from different
deep language models tested using two data augmentation techniques. Our
findings suggest developing a class of language models that infuses
psychological and clinical knowledge.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 23:52:35 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Garg",
"Muskan",
""
],
[
"Gaur",
"Manas",
""
],
[
"Goswami",
"Raxit",
""
],
[
"Sohn",
"Sunghwan",
""
]
] | TITLE: LOST: A Mental Health Dataset of Low Self-esteem in Reddit Posts
ABSTRACT: Low self-esteem and interpersonal needs (i.e., thwarted belongingness (TB)
and perceived burdensomeness (PB)) have a major impact on depression and
suicide attempts. Individuals seek social connectedness on social media to
boost and alleviate their loneliness. Social media platforms allow people to
express their thoughts, experiences, beliefs, and emotions. Prior studies on
mental health from social media have focused on symptoms, causes, and
disorders. Whereas an initial screening of social media content for
interpersonal risk factors and low self-esteem may raise early alerts and
assign therapists to at-risk users of mental disturbance. Standardized scales
measure self-esteem and interpersonal needs from questions created using
psychological theories. In the current research, we introduce a
psychology-grounded and expertly annotated dataset, LoST: Low Self esTeem, to
study and detect low self-esteem on Reddit. Through an annotation approach
involving checks on coherence, correctness, consistency, and reliability, we
ensure gold-standard for supervised learning. We present results from different
deep language models tested using two data augmentation techniques. Our
findings suggest developing a class of language models that infuses
psychological and clinical knowledge.
|
2306.05598 | Ilan Mandel | Madiha Zahrah Choksi, Ilan Mandel, David Goedicke, Yan Shvartzshnaider | Enclosed Loops: How open source communities become datasets | 9 pages | null | null | null | cs.CY | http://creativecommons.org/licenses/by-sa/4.0/ | Centralization in code hosting and package management in the 2010s created
fundamental shifts in the social arrangements of open source ecosystems. In a
regime of centralized open source, platform effects can both empower and
detract from communities depending on underlying technical implementations and
governance mechanisms. In this paper we examine Dependabot, Crater and Copilot
as three nascent tools whose existence is predicated on centralized software at
scale. Open source ecosystems are maintained by positive feedback loops between
community members and their outputs. This mechanism is guided by community
standards that foreground notions of accountability and transparency. On one
hand, software at scale supports positive feedback loops of exchange among
ecosystem stakeholders: community members (developers), users, and projects. On
the other, software at scale becomes a commodity to be leveraged and
expropriated. We perform a comparative analysis of attributes across the three
tools and evaluate their goals, values, and norms. We investigate these
feedback loops and their sociotechnical effects on open source communities. We
demonstrate how the values embedded in each case study may diverge from the
foundational ethos of open communities as they are motivated by, and respond to
the platform effects, corporate capture, and centralization of open source
infrastructure. Our analysis finds that these tools embed values that are
reflective of different modes of development - some are transparent and
accountable, and others are not. In doing so, certain tools may have feedback
mechanisms that extend communities. Others threaten and damage communities
ability to reproduce themselves.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 00:02:25 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Choksi",
"Madiha Zahrah",
""
],
[
"Mandel",
"Ilan",
""
],
[
"Goedicke",
"David",
""
],
[
"Shvartzshnaider",
"Yan",
""
]
] | TITLE: Enclosed Loops: How open source communities become datasets
ABSTRACT: Centralization in code hosting and package management in the 2010s created
fundamental shifts in the social arrangements of open source ecosystems. In a
regime of centralized open source, platform effects can both empower and
detract from communities depending on underlying technical implementations and
governance mechanisms. In this paper we examine Dependabot, Crater and Copilot
as three nascent tools whose existence is predicated on centralized software at
scale. Open source ecosystems are maintained by positive feedback loops between
community members and their outputs. This mechanism is guided by community
standards that foreground notions of accountability and transparency. On one
hand, software at scale supports positive feedback loops of exchange among
ecosystem stakeholders: community members (developers), users, and projects. On
the other, software at scale becomes a commodity to be leveraged and
expropriated. We perform a comparative analysis of attributes across the three
tools and evaluate their goals, values, and norms. We investigate these
feedback loops and their sociotechnical effects on open source communities. We
demonstrate how the values embedded in each case study may diverge from the
foundational ethos of open communities as they are motivated by, and respond to
the platform effects, corporate capture, and centralization of open source
infrastructure. Our analysis finds that these tools embed values that are
reflective of different modes of development - some are transparent and
accountable, and others are not. In doing so, certain tools may have feedback
mechanisms that extend communities. Others threaten and damage communities
ability to reproduce themselves.
|
2306.05605 | Keiji Shinzato | Keiji Shinzato, Naoki Yoshinaga, Yandi Xia and Wei-Te Chen | A Unified Generative Approach to Product Attribute-Value Identification | Accepted to the Findings of ACL 2023 | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Product attribute-value identification (PAVI) has been studied to link
products on e-commerce sites with their attribute values (e.g., <Material,
Cotton>) using product text as clues. Technical demands from real-world
e-commerce platforms require PAVI methods to handle unseen values,
multi-attribute values, and canonicalized values, which are only partly
addressed in existing extraction- and classification-based approaches.
Motivated by this, we explore a generative approach to the PAVI task. We
finetune a pre-trained generative model, T5, to decode a set of attribute-value
pairs as a target sequence from the given product text. Since the attribute
value pairs are unordered set elements, how to linearize them will matter; we,
thus, explore methods of composing an attribute-value pair and ordering the
pairs for the task. Experimental results confirm that our generation-based
approach outperforms the existing extraction and classification-based methods
on large-scale real-world datasets meant for those methods.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 00:33:30 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Shinzato",
"Keiji",
""
],
[
"Yoshinaga",
"Naoki",
""
],
[
"Xia",
"Yandi",
""
],
[
"Chen",
"Wei-Te",
""
]
] | TITLE: A Unified Generative Approach to Product Attribute-Value Identification
ABSTRACT: Product attribute-value identification (PAVI) has been studied to link
products on e-commerce sites with their attribute values (e.g., <Material,
Cotton>) using product text as clues. Technical demands from real-world
e-commerce platforms require PAVI methods to handle unseen values,
multi-attribute values, and canonicalized values, which are only partly
addressed in existing extraction- and classification-based approaches.
Motivated by this, we explore a generative approach to the PAVI task. We
finetune a pre-trained generative model, T5, to decode a set of attribute-value
pairs as a target sequence from the given product text. Since the attribute
value pairs are unordered set elements, how to linearize them will matter; we,
thus, explore methods of composing an attribute-value pair and ordering the
pairs for the task. Experimental results confirm that our generation-based
approach outperforms the existing extraction and classification-based methods
on large-scale real-world datasets meant for those methods.
|
2306.05622 | Mathias Weiden | Mathias Weiden, Ed Younis, Justin Kalloor, John Kubiatowicz, and
Costin Iancu | Improving Quantum Circuit Synthesis with Machine Learning | 11 pages, 10 figures | null | null | null | quant-ph cs.LG | http://creativecommons.org/licenses/by/4.0/ | In the Noisy Intermediate Scale Quantum (NISQ) era, finding implementations
of quantum algorithms that minimize the number of expensive and error prone
multi-qubit gates is vital to ensure computations produce meaningful outputs.
Unitary synthesis, the process of finding a quantum circuit that implements
some target unitary matrix, is able to solve this problem optimally in many
cases. However, current bottom-up unitary synthesis algorithms are limited by
their exponentially growing run times. We show how applying machine learning to
unitary datasets permits drastic speedups for synthesis algorithms. This paper
presents QSeed, a seeded synthesis algorithm that employs a learned model to
quickly propose resource efficient circuit implementations of unitaries. QSeed
maintains low gate counts and offers a speedup of $3.7\times$ in synthesis time
over the state of the art for a 64 qubit modular exponentiation circuit, a core
component in Shor's factoring algorithm. QSeed's performance improvements also
generalize to families of circuits not seen during the training process.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 01:53:56 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Weiden",
"Mathias",
""
],
[
"Younis",
"Ed",
""
],
[
"Kalloor",
"Justin",
""
],
[
"Kubiatowicz",
"John",
""
],
[
"Iancu",
"Costin",
""
]
] | TITLE: Improving Quantum Circuit Synthesis with Machine Learning
ABSTRACT: In the Noisy Intermediate Scale Quantum (NISQ) era, finding implementations
of quantum algorithms that minimize the number of expensive and error prone
multi-qubit gates is vital to ensure computations produce meaningful outputs.
Unitary synthesis, the process of finding a quantum circuit that implements
some target unitary matrix, is able to solve this problem optimally in many
cases. However, current bottom-up unitary synthesis algorithms are limited by
their exponentially growing run times. We show how applying machine learning to
unitary datasets permits drastic speedups for synthesis algorithms. This paper
presents QSeed, a seeded synthesis algorithm that employs a learned model to
quickly propose resource efficient circuit implementations of unitaries. QSeed
maintains low gate counts and offers a speedup of $3.7\times$ in synthesis time
over the state of the art for a 64 qubit modular exponentiation circuit, a core
component in Shor's factoring algorithm. QSeed's performance improvements also
generalize to families of circuits not seen during the training process.
|
2306.05624 | Yanxiong Li | Yufei Zeng, Yanxiong Li, Zhenfeng Zhou, Ruiqi Wang, Difeng Lu | Domestic Activities Classification from Audio Recordings Using
Multi-scale Dilated Depthwise Separable Convolutional Network | 5 pages, 2 figures, 4 tables. Accepted for publication in IEEE
MMSP2021 | null | null | null | eess.AS cs.SD | http://creativecommons.org/licenses/by-sa/4.0/ | Domestic activities classification (DAC) from audio recordings aims at
classifying audio recordings into pre-defined categories of domestic
activities, which is an effective way for estimation of daily activities
performed in home environment. In this paper, we propose a method for DAC from
audio recordings using a multi-scale dilated depthwise separable convolutional
network (DSCN). The DSCN is a lightweight neural network with small size of
parameters and thus suitable to be deployed in portable terminals with limited
computing resources. To expand the receptive field with the same size of DSCN's
parameters, dilated convolution, instead of normal convolution, is used in the
DSCN for further improving the DSCN's performance. In addition, the embeddings
of various scales learned by the dilated DSCN are concatenated as a multi-scale
embedding for representing property differences among various classes of
domestic activities. Evaluated on a public dataset of the Task 5 of the 2018
challenge on Detection and Classification of Acoustic Scenes and Events
(DCASE-2018), the results show that: both dilated convolution and multi-scale
embedding contribute to the performance improvement of the proposed method; and
the proposed method outperforms the methods based on state-of-the-art
lightweight network in terms of classification accuracy.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 02:15:33 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Zeng",
"Yufei",
""
],
[
"Li",
"Yanxiong",
""
],
[
"Zhou",
"Zhenfeng",
""
],
[
"Wang",
"Ruiqi",
""
],
[
"Lu",
"Difeng",
""
]
] | TITLE: Domestic Activities Classification from Audio Recordings Using
Multi-scale Dilated Depthwise Separable Convolutional Network
ABSTRACT: Domestic activities classification (DAC) from audio recordings aims at
classifying audio recordings into pre-defined categories of domestic
activities, which is an effective way for estimation of daily activities
performed in home environment. In this paper, we propose a method for DAC from
audio recordings using a multi-scale dilated depthwise separable convolutional
network (DSCN). The DSCN is a lightweight neural network with small size of
parameters and thus suitable to be deployed in portable terminals with limited
computing resources. To expand the receptive field with the same size of DSCN's
parameters, dilated convolution, instead of normal convolution, is used in the
DSCN for further improving the DSCN's performance. In addition, the embeddings
of various scales learned by the dilated DSCN are concatenated as a multi-scale
embedding for representing property differences among various classes of
domestic activities. Evaluated on a public dataset of the Task 5 of the 2018
challenge on Detection and Classification of Acoustic Scenes and Events
(DCASE-2018), the results show that: both dilated convolution and multi-scale
embedding contribute to the performance improvement of the proposed method; and
the proposed method outperforms the methods based on state-of-the-art
lightweight network in terms of classification accuracy.
|
2306.05628 | Lirong Wu | Lirong Wu, Haitao Lin, Yufei Huang, Stan Z. Li | Quantifying the Knowledge in GNNs for Reliable Distillation into MLPs | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | To bridge the gaps between topology-aware Graph Neural Networks (GNNs) and
inference-efficient Multi-Layer Perceptron (MLPs), GLNN proposes to distill
knowledge from a well-trained teacher GNN into a student MLP. Despite their
great progress, comparatively little work has been done to explore the
reliability of different knowledge points (nodes) in GNNs, especially their
roles played during distillation. In this paper, we first quantify the
knowledge reliability in GNN by measuring the invariance of their information
entropy to noise perturbations, from which we observe that different knowledge
points (1) show different distillation speeds (temporally); (2) are
differentially distributed in the graph (spatially). To achieve reliable
distillation, we propose an effective approach, namely Knowledge-inspired
Reliable Distillation (KRD), that models the probability of each node being an
informative and reliable knowledge point, based on which we sample a set of
additional reliable knowledge points as supervision for training student MLPs.
Extensive experiments show that KRD improves over the vanilla MLPs by 12.62%
and outperforms its corresponding teacher GNNs by 2.16% averaged over 7
datasets and 3 GNN architectures.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 02:23:37 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wu",
"Lirong",
""
],
[
"Lin",
"Haitao",
""
],
[
"Huang",
"Yufei",
""
],
[
"Li",
"Stan Z.",
""
]
] | TITLE: Quantifying the Knowledge in GNNs for Reliable Distillation into MLPs
ABSTRACT: To bridge the gaps between topology-aware Graph Neural Networks (GNNs) and
inference-efficient Multi-Layer Perceptron (MLPs), GLNN proposes to distill
knowledge from a well-trained teacher GNN into a student MLP. Despite their
great progress, comparatively little work has been done to explore the
reliability of different knowledge points (nodes) in GNNs, especially their
roles played during distillation. In this paper, we first quantify the
knowledge reliability in GNN by measuring the invariance of their information
entropy to noise perturbations, from which we observe that different knowledge
points (1) show different distillation speeds (temporally); (2) are
differentially distributed in the graph (spatially). To achieve reliable
distillation, we propose an effective approach, namely Knowledge-inspired
Reliable Distillation (KRD), that models the probability of each node being an
informative and reliable knowledge point, based on which we sample a set of
additional reliable knowledge points as supervision for training student MLPs.
Extensive experiments show that KRD improves over the vanilla MLPs by 12.62%
and outperforms its corresponding teacher GNNs by 2.16% averaged over 7
datasets and 3 GNN architectures.
|
2306.05636 | Armin Toroghi | Armin Toroghi, Griffin Floto, Zhenwei Tang, Scott Sanner | Bayesian Knowledge-driven Critiquing with Indirect Evidence | null | null | 10.1145/3539618.3591954 | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conversational recommender systems (CRS) enhance the expressivity and
personalization of recommendations through multiple turns of user-system
interaction. Critiquing is a well-known paradigm for CRS that allows users to
iteratively refine recommendations by providing feedback about attributes of
recommended items. While existing critiquing methodologies utilize direct
attributes of items to address user requests such as 'I prefer Western movies',
the opportunity of incorporating richer contextual and side information about
items stored in Knowledge Graphs (KG) into the critiquing paradigm has been
overlooked. Employing this substantial knowledge together with a
well-established reasoning methodology paves the way for critique-based
recommenders to allow for complex knowledge-based feedback (e.g., 'I like
movies featuring war side effects on veterans') which may arise in natural
user-system conversations. In this work, we aim to increase the flexibility of
critique-based recommendation by integrating KGs and propose a novel Bayesian
inference framework that enables reasoning with relational knowledge-based
feedback. We study and formulate the framework considering a Gaussian
likelihood and evaluate it on two well-known recommendation datasets with KGs.
Our evaluations demonstrate the effectiveness of our framework in leveraging
indirect KG-based feedback (i.e., preferred relational properties of items
rather than preferred items themselves), often improving personalized
recommendations over a one-shot recommender by more than 15%. This work enables
a new paradigm for using rich knowledge content and reasoning over indirect
evidence as a mechanism for critiquing interactions with CRS.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 02:43:26 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Toroghi",
"Armin",
""
],
[
"Floto",
"Griffin",
""
],
[
"Tang",
"Zhenwei",
""
],
[
"Sanner",
"Scott",
""
]
] | TITLE: Bayesian Knowledge-driven Critiquing with Indirect Evidence
ABSTRACT: Conversational recommender systems (CRS) enhance the expressivity and
personalization of recommendations through multiple turns of user-system
interaction. Critiquing is a well-known paradigm for CRS that allows users to
iteratively refine recommendations by providing feedback about attributes of
recommended items. While existing critiquing methodologies utilize direct
attributes of items to address user requests such as 'I prefer Western movies',
the opportunity of incorporating richer contextual and side information about
items stored in Knowledge Graphs (KG) into the critiquing paradigm has been
overlooked. Employing this substantial knowledge together with a
well-established reasoning methodology paves the way for critique-based
recommenders to allow for complex knowledge-based feedback (e.g., 'I like
movies featuring war side effects on veterans') which may arise in natural
user-system conversations. In this work, we aim to increase the flexibility of
critique-based recommendation by integrating KGs and propose a novel Bayesian
inference framework that enables reasoning with relational knowledge-based
feedback. We study and formulate the framework considering a Gaussian
likelihood and evaluate it on two well-known recommendation datasets with KGs.
Our evaluations demonstrate the effectiveness of our framework in leveraging
indirect KG-based feedback (i.e., preferred relational properties of items
rather than preferred items themselves), often improving personalized
recommendations over a one-shot recommender by more than 15%. This work enables
a new paradigm for using rich knowledge content and reasoning over indirect
evidence as a mechanism for critiquing interactions with CRS.
|
2306.05637 | Hojoon Lee | Hojoon Lee and Koanho Lee and Dongyoon Hwang and Hyunho Lee and
Byungkun Lee and Jaegul Choo | On the Importance of Feature Decorrelation for Unsupervised
Representation Learning in Reinforcement Learning | Accepted to ICML 2023 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Recently, unsupervised representation learning (URL) has improved the sample
efficiency of Reinforcement Learning (RL) by pretraining a model from a large
unlabeled dataset. The underlying principle of these methods is to learn
temporally predictive representations by predicting future states in the latent
space. However, an important challenge of this approach is the representational
collapse, where the subspace of the latent representations collapses into a
low-dimensional manifold. To address this issue, we propose a novel URL
framework that causally predicts future states while increasing the dimension
of the latent manifold by decorrelating the features in the latent space.
Through extensive empirical studies, we demonstrate that our framework
effectively learns predictive representations without collapse, which
significantly improves the sample efficiency of state-of-the-art URL methods on
the Atari 100k benchmark. The code is available at
https://github.com/dojeon-ai/SimTPR.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 02:47:21 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lee",
"Hojoon",
""
],
[
"Lee",
"Koanho",
""
],
[
"Hwang",
"Dongyoon",
""
],
[
"Lee",
"Hyunho",
""
],
[
"Lee",
"Byungkun",
""
],
[
"Choo",
"Jaegul",
""
]
] | TITLE: On the Importance of Feature Decorrelation for Unsupervised
Representation Learning in Reinforcement Learning
ABSTRACT: Recently, unsupervised representation learning (URL) has improved the sample
efficiency of Reinforcement Learning (RL) by pretraining a model from a large
unlabeled dataset. The underlying principle of these methods is to learn
temporally predictive representations by predicting future states in the latent
space. However, an important challenge of this approach is the representational
collapse, where the subspace of the latent representations collapses into a
low-dimensional manifold. To address this issue, we propose a novel URL
framework that causally predicts future states while increasing the dimension
of the latent manifold by decorrelating the features in the latent space.
Through extensive empirical studies, we demonstrate that our framework
effectively learns predictive representations without collapse, which
significantly improves the sample efficiency of state-of-the-art URL methods on
the Atari 100k benchmark. The code is available at
https://github.com/dojeon-ai/SimTPR.
|
2306.05652 | Prateek Chhikara | Prateek Chhikara, Ujjwal Pasupulety, John Marshall, Dhiraj Chaurasia,
Shweta Kumari | Privacy Aware Question-Answering System for Online Mental Health Risk
Assessment | 5 pages, 2 figures, 3 tables | null | null | null | cs.CL cs.AI cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Social media platforms have enabled individuals suffering from mental
illnesses to share their lived experiences and find the online support
necessary to cope. However, many users fail to receive genuine clinical
support, thus exacerbating their symptoms. Screening users based on what they
post online can aid providers in administering targeted healthcare and minimize
false positives. Pre-trained Language Models (LMs) can assess users' social
media data and classify them in terms of their mental health risk. We propose a
Question-Answering (QA) approach to assess mental health risk using the
Unified-QA model on two large mental health datasets. To protect user data, we
extend Unified-QA by anonymizing the model training process using differential
privacy. Our results demonstrate the effectiveness of modeling risk assessment
as a QA task, specifically for mental health use cases. Furthermore, the
model's performance decreases by less than 1% with the inclusion of
differential privacy. The proposed system's performance is indicative of a
promising research direction that will lead to the development of privacy-aware
diagnostic systems.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 03:37:49 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Chhikara",
"Prateek",
""
],
[
"Pasupulety",
"Ujjwal",
""
],
[
"Marshall",
"John",
""
],
[
"Chaurasia",
"Dhiraj",
""
],
[
"Kumari",
"Shweta",
""
]
] | TITLE: Privacy Aware Question-Answering System for Online Mental Health Risk
Assessment
ABSTRACT: Social media platforms have enabled individuals suffering from mental
illnesses to share their lived experiences and find the online support
necessary to cope. However, many users fail to receive genuine clinical
support, thus exacerbating their symptoms. Screening users based on what they
post online can aid providers in administering targeted healthcare and minimize
false positives. Pre-trained Language Models (LMs) can assess users' social
media data and classify them in terms of their mental health risk. We propose a
Question-Answering (QA) approach to assess mental health risk using the
Unified-QA model on two large mental health datasets. To protect user data, we
extend Unified-QA by anonymizing the model training process using differential
privacy. Our results demonstrate the effectiveness of modeling risk assessment
as a QA task, specifically for mental health use cases. Furthermore, the
model's performance decreases by less than 1% with the inclusion of
differential privacy. The proposed system's performance is indicative of a
promising research direction that will lead to the development of privacy-aware
diagnostic systems.
|
2306.05663 | Eduardo R. Corral-Soto | Eduardo R. Corral-Soto, Alaap Grandhi, Yannis Y. He, Mrigank Rochan,
Bingbing Liu | Improving LiDAR 3D Object Detection via Range-based Point Cloud Density
Optimization | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, much progress has been made in LiDAR-based 3D object
detection mainly due to advances in detector architecture designs and
availability of large-scale LiDAR datasets. Existing 3D object detectors tend
to perform well on the point cloud regions closer to the LiDAR sensor as
opposed to on regions that are farther away. In this paper, we investigate this
problem from the data perspective instead of detector architecture design. We
observe that there is a learning bias in detection models towards the dense
objects near the sensor and show that the detection performance can be improved
by simply manipulating the input point cloud density at different distance
ranges without modifying the detector architecture and without data
augmentation. We propose a model-free point cloud density adjustment
pre-processing mechanism that uses iterative MCMC optimization to estimate
optimal parameters for altering the point density at different distance ranges.
We conduct experiments using four state-of-the-art LiDAR 3D object detectors on
two public LiDAR datasets, namely Waymo and ONCE. Our results demonstrate that
our range-based point cloud density manipulation technique can improve the
performance of the existing detectors, which in turn could potentially inspire
future detector designs.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 04:11:43 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Corral-Soto",
"Eduardo R.",
""
],
[
"Grandhi",
"Alaap",
""
],
[
"He",
"Yannis Y.",
""
],
[
"Rochan",
"Mrigank",
""
],
[
"Liu",
"Bingbing",
""
]
] | TITLE: Improving LiDAR 3D Object Detection via Range-based Point Cloud Density
Optimization
ABSTRACT: In recent years, much progress has been made in LiDAR-based 3D object
detection mainly due to advances in detector architecture designs and
availability of large-scale LiDAR datasets. Existing 3D object detectors tend
to perform well on the point cloud regions closer to the LiDAR sensor as
opposed to on regions that are farther away. In this paper, we investigate this
problem from the data perspective instead of detector architecture design. We
observe that there is a learning bias in detection models towards the dense
objects near the sensor and show that the detection performance can be improved
by simply manipulating the input point cloud density at different distance
ranges without modifying the detector architecture and without data
augmentation. We propose a model-free point cloud density adjustment
pre-processing mechanism that uses iterative MCMC optimization to estimate
optimal parameters for altering the point density at different distance ranges.
We conduct experiments using four state-of-the-art LiDAR 3D object detectors on
two public LiDAR datasets, namely Waymo and ONCE. Our results demonstrate that
our range-based point cloud density manipulation technique can improve the
performance of the existing detectors, which in turn could potentially inspire
future detector designs.
|
2306.05682 | Dong-Jae Lee | Dong-Jae Lee, Jae Young Lee, Hyounguk Shon, Eojindl Yi, Yeong-Hun
Park, Sung-Sik Cho, Junmo Kim | Lightweight Monocular Depth Estimation via Token-Sharing Transformer | ICRA 2023 | null | null | null | cs.CV cs.AI cs.LG cs.RO eess.IV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Depth estimation is an important task in various robotics systems and
applications. In mobile robotics systems, monocular depth estimation is
desirable since a single RGB camera can be deployable at a low cost and compact
size. Due to its significant and growing needs, many lightweight monocular
depth estimation networks have been proposed for mobile robotics systems. While
most lightweight monocular depth estimation methods have been developed using
convolution neural networks, the Transformer has been gradually utilized in
monocular depth estimation recently. However, massive parameters and large
computational costs in the Transformer disturb the deployment to embedded
devices. In this paper, we present a Token-Sharing Transformer (TST), an
architecture using the Transformer for monocular depth estimation, optimized
especially in embedded devices. The proposed TST utilizes global token sharing,
which enables the model to obtain an accurate depth prediction with high
throughput in embedded devices. Experimental results show that TST outperforms
the existing lightweight monocular depth estimation methods. On the NYU Depth
v2 dataset, TST can deliver depth maps up to 63.4 FPS in NVIDIA Jetson nano and
142.6 FPS in NVIDIA Jetson TX2, with lower errors than the existing methods.
Furthermore, TST achieves real-time depth estimation of high-resolution images
on Jetson TX2 with competitive results.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 05:51:40 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Lee",
"Dong-Jae",
""
],
[
"Lee",
"Jae Young",
""
],
[
"Shon",
"Hyounguk",
""
],
[
"Yi",
"Eojindl",
""
],
[
"Park",
"Yeong-Hun",
""
],
[
"Cho",
"Sung-Sik",
""
],
[
"Kim",
"Junmo",
""
]
] | TITLE: Lightweight Monocular Depth Estimation via Token-Sharing Transformer
ABSTRACT: Depth estimation is an important task in various robotics systems and
applications. In mobile robotics systems, monocular depth estimation is
desirable since a single RGB camera can be deployable at a low cost and compact
size. Due to its significant and growing needs, many lightweight monocular
depth estimation networks have been proposed for mobile robotics systems. While
most lightweight monocular depth estimation methods have been developed using
convolution neural networks, the Transformer has been gradually utilized in
monocular depth estimation recently. However, massive parameters and large
computational costs in the Transformer disturb the deployment to embedded
devices. In this paper, we present a Token-Sharing Transformer (TST), an
architecture using the Transformer for monocular depth estimation, optimized
especially in embedded devices. The proposed TST utilizes global token sharing,
which enables the model to obtain an accurate depth prediction with high
throughput in embedded devices. Experimental results show that TST outperforms
the existing lightweight monocular depth estimation methods. On the NYU Depth
v2 dataset, TST can deliver depth maps up to 63.4 FPS in NVIDIA Jetson nano and
142.6 FPS in NVIDIA Jetson TX2, with lower errors than the existing methods.
Furthermore, TST achieves real-time depth estimation of high-resolution images
on Jetson TX2 with competitive results.
|
2306.05689 | Wu Tz-Ying | Alakh Desai, Tz-Ying Wu, Subarna Tripathi, Nuno Vasconcelos | Single-Stage Visual Relationship Learning using Conditional Queries | Accepted to NeurIPS 2022 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Research in scene graph generation (SGG) usually considers two-stage models,
that is, detecting a set of entities, followed by combining them and labeling
all possible relationships. While showing promising results, the pipeline
structure induces large parameter and computation overhead, and typically
hinders end-to-end optimizations. To address this, recent research attempts to
train single-stage models that are computationally efficient. With the advent
of DETR, a set based detection model, one-stage models attempt to predict a set
of subject-predicate-object triplets directly in a single shot. However, SGG is
inherently a multi-task learning problem that requires modeling entity and
predicate distributions simultaneously. In this paper, we propose Transformers
with conditional queries for SGG, namely, TraCQ with a new formulation for SGG
that avoids the multi-task learning problem and the combinatorial entity pair
distribution. We employ a DETR-based encoder-decoder design and leverage
conditional queries to significantly reduce the entity label space as well,
which leads to 20% fewer parameters compared to state-of-the-art single-stage
models. Experimental results show that TraCQ not only outperforms existing
single-stage scene graph generation methods, it also beats many
state-of-the-art two-stage methods on the Visual Genome dataset, yet is capable
of end-to-end training and faster inference.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 06:02:01 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Desai",
"Alakh",
""
],
[
"Wu",
"Tz-Ying",
""
],
[
"Tripathi",
"Subarna",
""
],
[
"Vasconcelos",
"Nuno",
""
]
] | TITLE: Single-Stage Visual Relationship Learning using Conditional Queries
ABSTRACT: Research in scene graph generation (SGG) usually considers two-stage models,
that is, detecting a set of entities, followed by combining them and labeling
all possible relationships. While showing promising results, the pipeline
structure induces large parameter and computation overhead, and typically
hinders end-to-end optimizations. To address this, recent research attempts to
train single-stage models that are computationally efficient. With the advent
of DETR, a set based detection model, one-stage models attempt to predict a set
of subject-predicate-object triplets directly in a single shot. However, SGG is
inherently a multi-task learning problem that requires modeling entity and
predicate distributions simultaneously. In this paper, we propose Transformers
with conditional queries for SGG, namely, TraCQ with a new formulation for SGG
that avoids the multi-task learning problem and the combinatorial entity pair
distribution. We employ a DETR-based encoder-decoder design and leverage
conditional queries to significantly reduce the entity label space as well,
which leads to 20% fewer parameters compared to state-of-the-art single-stage
models. Experimental results show that TraCQ not only outperforms existing
single-stage scene graph generation methods, it also beats many
state-of-the-art two-stage methods on the Visual Genome dataset, yet is capable
of end-to-end training and faster inference.
|
2306.05695 | Yinghui Ye | Haohang Yang, Yinghui Ye, Kai Liang, Xiaoli Chu | Power Beacon Energy Consumption Minimization in Wireless Powered
Backscatter Communication Networks | null | null | null | null | cs.IT eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Internet-of-Things (IoT) networks are expected to support the wireless
connection of massive energy limited IoT nodes. The emerging wireless powered
backscatter communications (WPBC) enable IoT nodes to harvest energy from the
incident radio frequency signals transmitted by a power beacon (PB) to support
their circuit operation, but the energy consumption of the PB (a potentially
high cost borne by the network operator) has not been sufficiently studied for
WPBC. In this paper, we aim to minimize the energy consumption of the PB while
satisfying the throughput requirement per IoT node by jointly optimizing the
time division multiple access (TDMA) time slot duration and backscatter
reflection coefficient of each IoT node and the PB transmit power per time
slot. As the formulated joint optimization problem is non-convex, we transform
it into a convex problem by using auxiliary variables, then employ the Lagrange
dual method to obtain the optimal solutions. To reduce the implementation
complexity required for adjusting the PB's transmit power every time slot, we
keep the PB transmit power constant in each time block and solve the
corresponding PB energy consumption minimization problem by using auxiliary
variables, the block coordinated decent method and the successive convex
approximation technique. Based on the above solutions, two iterative algorithms
are proposed for the dynamic PB transmit power scheme and the static PB
transmit power scheme. The simulation results show that the dynamic PB transmit
power scheme and the static PB transmit power scheme both achieve a lower PB
energy consumption than the benchmark schemes, and the former achieves the
lowest PB energy consumption.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 06:33:27 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Yang",
"Haohang",
""
],
[
"Ye",
"Yinghui",
""
],
[
"Liang",
"Kai",
""
],
[
"Chu",
"Xiaoli",
""
]
] | TITLE: Power Beacon Energy Consumption Minimization in Wireless Powered
Backscatter Communication Networks
ABSTRACT: Internet-of-Things (IoT) networks are expected to support the wireless
connection of massive energy limited IoT nodes. The emerging wireless powered
backscatter communications (WPBC) enable IoT nodes to harvest energy from the
incident radio frequency signals transmitted by a power beacon (PB) to support
their circuit operation, but the energy consumption of the PB (a potentially
high cost borne by the network operator) has not been sufficiently studied for
WPBC. In this paper, we aim to minimize the energy consumption of the PB while
satisfying the throughput requirement per IoT node by jointly optimizing the
time division multiple access (TDMA) time slot duration and backscatter
reflection coefficient of each IoT node and the PB transmit power per time
slot. As the formulated joint optimization problem is non-convex, we transform
it into a convex problem by using auxiliary variables, then employ the Lagrange
dual method to obtain the optimal solutions. To reduce the implementation
complexity required for adjusting the PB's transmit power every time slot, we
keep the PB transmit power constant in each time block and solve the
corresponding PB energy consumption minimization problem by using auxiliary
variables, the block coordinated decent method and the successive convex
approximation technique. Based on the above solutions, two iterative algorithms
are proposed for the dynamic PB transmit power scheme and the static PB
transmit power scheme. The simulation results show that the dynamic PB transmit
power scheme and the static PB transmit power scheme both achieve a lower PB
energy consumption than the benchmark schemes, and the former achieves the
lowest PB energy consumption.
|
2306.05698 | Naoto Yanai | Chika Komiya and Naoto Yanai and Kyosuke Yamashita and Shingo Okamura | JABBERWOCK: A Tool for WebAssembly Dataset Generation and Its
Application to Malicious Website Detection | Accepted in DCDS 2023 (co-located in DSN 2023) | null | null | null | cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Machine learning is often used for malicious website detection, but an
approach incorporating WebAssembly as a feature has not been explored due to a
limited number of samples, to the best of our knowledge. In this paper, we
propose JABBERWOCK (JAvascript-Based Binary EncodeR by WebAssembly Optimization
paCKer), a tool to generate WebAssembly datasets in a pseudo fashion via
JavaScript. Loosely speaking, JABBERWOCK automatically gathers JavaScript code
in the real world, convert them into WebAssembly, and then outputs vectors of
the WebAssembly as samples for malicious website detection. We also conduct
experimental evaluations of JABBERWOCK in terms of the processing time for
dataset generation, comparison of the generated samples with actual WebAssembly
samples gathered from the Internet, and an application for malicious website
detection. Regarding the processing time, we show that JABBERWOCK can construct
a dataset in 4.5 seconds per sample for any number of samples. Next, comparing
10,000 samples output by JABBERWOCK with 168 gathered WebAssembly samples, we
believe that the generated samples by JABBERWOCK are similar to those in the
real world. We then show that JABBERWOCK can provide malicious website
detection with 99\% F1-score because JABBERWOCK makes a gap between benign and
malicious samples as the reason for the above high score. We also confirm that
JABBERWOCK can be combined with an existing malicious website detection tool to
improve F1-scores. JABBERWOCK is publicly available via GitHub
(https://github.com/c-chocolate/Jabberwock).
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 06:35:14 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Komiya",
"Chika",
""
],
[
"Yanai",
"Naoto",
""
],
[
"Yamashita",
"Kyosuke",
""
],
[
"Okamura",
"Shingo",
""
]
] | TITLE: JABBERWOCK: A Tool for WebAssembly Dataset Generation and Its
Application to Malicious Website Detection
ABSTRACT: Machine learning is often used for malicious website detection, but an
approach incorporating WebAssembly as a feature has not been explored due to a
limited number of samples, to the best of our knowledge. In this paper, we
propose JABBERWOCK (JAvascript-Based Binary EncodeR by WebAssembly Optimization
paCKer), a tool to generate WebAssembly datasets in a pseudo fashion via
JavaScript. Loosely speaking, JABBERWOCK automatically gathers JavaScript code
in the real world, convert them into WebAssembly, and then outputs vectors of
the WebAssembly as samples for malicious website detection. We also conduct
experimental evaluations of JABBERWOCK in terms of the processing time for
dataset generation, comparison of the generated samples with actual WebAssembly
samples gathered from the Internet, and an application for malicious website
detection. Regarding the processing time, we show that JABBERWOCK can construct
a dataset in 4.5 seconds per sample for any number of samples. Next, comparing
10,000 samples output by JABBERWOCK with 168 gathered WebAssembly samples, we
believe that the generated samples by JABBERWOCK are similar to those in the
real world. We then show that JABBERWOCK can provide malicious website
detection with 99\% F1-score because JABBERWOCK makes a gap between benign and
malicious samples as the reason for the above high score. We also confirm that
JABBERWOCK can be combined with an existing malicious website detection tool to
improve F1-scores. JABBERWOCK is publicly available via GitHub
(https://github.com/c-chocolate/Jabberwock).
|
2306.05704 | Lin Liu | Lin Liu, Mingming Zhao, Shanxin Yuan, Wenlong Lyu, Wengang Zhou,
Houqiang Li, Yanfeng Wang, Qi Tian | Exploring Effective Mask Sampling Modeling for Neural Image Compression | 10 pages | null | null | null | cs.CV cs.MM eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image compression aims to reduce the information redundancy in images. Most
existing neural image compression methods rely on side information from
hyperprior or context models to eliminate spatial redundancy, but rarely
address the channel redundancy. Inspired by the mask sampling modeling in
recent self-supervised learning methods for natural language processing and
high-level vision, we propose a novel pretraining strategy for neural image
compression. Specifically, Cube Mask Sampling Module (CMSM) is proposed to
apply both spatial and channel mask sampling modeling to image compression in
the pre-training stage. Moreover, to further reduce channel redundancy, we
propose the Learnable Channel Mask Module (LCMM) and the Learnable Channel
Completion Module (LCCM). Our plug-and-play CMSM, LCMM, LCCM modules can apply
to both CNN-based and Transformer-based architectures, significantly reduce the
computational cost, and improve the quality of images. Experiments on the
public Kodak and Tecnick datasets demonstrate that our method achieves
competitive performance with lower computational complexity compared to
state-of-the-art image compression methods.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 06:50:20 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Liu",
"Lin",
""
],
[
"Zhao",
"Mingming",
""
],
[
"Yuan",
"Shanxin",
""
],
[
"Lyu",
"Wenlong",
""
],
[
"Zhou",
"Wengang",
""
],
[
"Li",
"Houqiang",
""
],
[
"Wang",
"Yanfeng",
""
],
[
"Tian",
"Qi",
""
]
] | TITLE: Exploring Effective Mask Sampling Modeling for Neural Image Compression
ABSTRACT: Image compression aims to reduce the information redundancy in images. Most
existing neural image compression methods rely on side information from
hyperprior or context models to eliminate spatial redundancy, but rarely
address the channel redundancy. Inspired by the mask sampling modeling in
recent self-supervised learning methods for natural language processing and
high-level vision, we propose a novel pretraining strategy for neural image
compression. Specifically, Cube Mask Sampling Module (CMSM) is proposed to
apply both spatial and channel mask sampling modeling to image compression in
the pre-training stage. Moreover, to further reduce channel redundancy, we
propose the Learnable Channel Mask Module (LCMM) and the Learnable Channel
Completion Module (LCCM). Our plug-and-play CMSM, LCMM, LCCM modules can apply
to both CNN-based and Transformer-based architectures, significantly reduce the
computational cost, and improve the quality of images. Experiments on the
public Kodak and Tecnick datasets demonstrate that our method achieves
competitive performance with lower computational complexity compared to
state-of-the-art image compression methods.
|
2306.05706 | Yan Sun | Yan Sun, Li Shen, Dacheng Tao | Understanding How Consistency Works in Federated Learning via Stage-wise
Relaxed Initialization | 32 pages | null | null | null | cs.LG cs.DC math.OC | http://creativecommons.org/licenses/by/4.0/ | Federated learning (FL) is a distributed paradigm that coordinates massive
local clients to collaboratively train a global model via stage-wise local
training processes on the heterogeneous dataset. Previous works have implicitly
studied that FL suffers from the ``client-drift'' problem, which is caused by
the inconsistent optimum across local clients. However, till now it still lacks
solid theoretical analysis to explain the impact of this local inconsistency.
To alleviate the negative impact of the ``client drift'' and explore its
substance in FL, in this paper, we first design an efficient FL algorithm
\textit{FedInit}, which allows employing the personalized relaxed
initialization state at the beginning of each local training stage.
Specifically, \textit{FedInit} initializes the local state by moving away from
the current global state towards the reverse direction of the latest local
state. This relaxed initialization helps to revise the local divergence and
enhance the local consistency level. Moreover, to further understand how
inconsistency disrupts performance in FL, we introduce the excess risk analysis
and study the divergence term to investigate the test error of the proposed
\textit{FedInit} method. Our studies show that optimization error is not
sensitive to this local inconsistency, while it mainly affects the
generalization error bound in \textit{FedInit}. Extensive experiments are
conducted to validate this conclusion. Our proposed \textit{FedInit} could
achieve state-of-the-art~(SOTA) results compared to several advanced benchmarks
without any additional costs. Meanwhile, stage-wise relaxed initialization
could also be incorporated into the current advanced algorithms to achieve
higher performance in the FL paradigm.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 06:55:15 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Sun",
"Yan",
""
],
[
"Shen",
"Li",
""
],
[
"Tao",
"Dacheng",
""
]
] | TITLE: Understanding How Consistency Works in Federated Learning via Stage-wise
Relaxed Initialization
ABSTRACT: Federated learning (FL) is a distributed paradigm that coordinates massive
local clients to collaboratively train a global model via stage-wise local
training processes on the heterogeneous dataset. Previous works have implicitly
studied that FL suffers from the ``client-drift'' problem, which is caused by
the inconsistent optimum across local clients. However, till now it still lacks
solid theoretical analysis to explain the impact of this local inconsistency.
To alleviate the negative impact of the ``client drift'' and explore its
substance in FL, in this paper, we first design an efficient FL algorithm
\textit{FedInit}, which allows employing the personalized relaxed
initialization state at the beginning of each local training stage.
Specifically, \textit{FedInit} initializes the local state by moving away from
the current global state towards the reverse direction of the latest local
state. This relaxed initialization helps to revise the local divergence and
enhance the local consistency level. Moreover, to further understand how
inconsistency disrupts performance in FL, we introduce the excess risk analysis
and study the divergence term to investigate the test error of the proposed
\textit{FedInit} method. Our studies show that optimization error is not
sensitive to this local inconsistency, while it mainly affects the
generalization error bound in \textit{FedInit}. Extensive experiments are
conducted to validate this conclusion. Our proposed \textit{FedInit} could
achieve state-of-the-art~(SOTA) results compared to several advanced benchmarks
without any additional costs. Meanwhile, stage-wise relaxed initialization
could also be incorporated into the current advanced algorithms to achieve
higher performance in the FL paradigm.
|
2306.05709 | Shijun Wang | Shijun Wang, J\'on Gu{\dh}nason, Damian Borth | Learning Emotional Representations from Imbalanced Speech Data for
Speech Emotion Recognition and Emotional Text-to-Speech | Accepted by INTERSPEECH2023 | null | null | null | eess.AS cs.CL cs.SD | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Effective speech emotional representations play a key role in Speech Emotion
Recognition (SER) and Emotional Text-To-Speech (TTS) tasks. However, emotional
speech samples are more difficult and expensive to acquire compared with
Neutral style speech, which causes one issue that most related works
unfortunately neglect: imbalanced datasets. Models might overfit to the
majority Neutral class and fail to produce robust and effective emotional
representations. In this paper, we propose an Emotion Extractor to address this
issue. We use augmentation approaches to train the model and enable it to
extract effective and generalizable emotional representations from imbalanced
datasets. Our empirical results show that (1) for the SER task, the proposed
Emotion Extractor surpasses the state-of-the-art baseline on three imbalanced
datasets; (2) the produced representations from our Emotion Extractor benefit
the TTS model, and enable it to synthesize more expressive speech.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 07:04:56 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wang",
"Shijun",
""
],
[
"Guðnason",
"Jón",
""
],
[
"Borth",
"Damian",
""
]
] | TITLE: Learning Emotional Representations from Imbalanced Speech Data for
Speech Emotion Recognition and Emotional Text-to-Speech
ABSTRACT: Effective speech emotional representations play a key role in Speech Emotion
Recognition (SER) and Emotional Text-To-Speech (TTS) tasks. However, emotional
speech samples are more difficult and expensive to acquire compared with
Neutral style speech, which causes one issue that most related works
unfortunately neglect: imbalanced datasets. Models might overfit to the
majority Neutral class and fail to produce robust and effective emotional
representations. In this paper, we propose an Emotion Extractor to address this
issue. We use augmentation approaches to train the model and enable it to
extract effective and generalizable emotional representations from imbalanced
datasets. Our empirical results show that (1) for the SER task, the proposed
Emotion Extractor surpasses the state-of-the-art baseline on three imbalanced
datasets; (2) the produced representations from our Emotion Extractor benefit
the TTS model, and enable it to synthesize more expressive speech.
|
2306.05747 | Pierre Tassel | Pierre Tassel, Martin Gebser, Konstantin Schekotihin | An End-to-End Reinforcement Learning Approach for Job-Shop Scheduling
Problems Based on Constraint Programming | To be published at ICAPS 2023 | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Constraint Programming (CP) is a declarative programming paradigm that allows
for modeling and solving combinatorial optimization problems, such as the
Job-Shop Scheduling Problem (JSSP). While CP solvers manage to find optimal or
near-optimal solutions for small instances, they do not scale well to large
ones, i.e., they require long computation times or yield low-quality solutions.
Therefore, real-world scheduling applications often resort to fast,
handcrafted, priority-based dispatching heuristics to find a good initial
solution and then refine it using optimization methods.
This paper proposes a novel end-to-end approach to solving scheduling
problems by means of CP and Reinforcement Learning (RL). In contrast to
previous RL methods, tailored for a given problem by including procedural
simulation algorithms, complex feature engineering, or handcrafted reward
functions, our neural-network architecture and training algorithm merely
require a generic CP encoding of some scheduling problem along with a set of
small instances. Our approach leverages existing CP solvers to train an agent
learning a Priority Dispatching Rule (PDR) that generalizes well to large
instances, even from separate datasets. We evaluate our method on seven JSSP
datasets from the literature, showing its ability to find higher-quality
solutions for very large instances than obtained by static PDRs and by a CP
solver within the same time limit.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 08:24:56 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Tassel",
"Pierre",
""
],
[
"Gebser",
"Martin",
""
],
[
"Schekotihin",
"Konstantin",
""
]
] | TITLE: An End-to-End Reinforcement Learning Approach for Job-Shop Scheduling
Problems Based on Constraint Programming
ABSTRACT: Constraint Programming (CP) is a declarative programming paradigm that allows
for modeling and solving combinatorial optimization problems, such as the
Job-Shop Scheduling Problem (JSSP). While CP solvers manage to find optimal or
near-optimal solutions for small instances, they do not scale well to large
ones, i.e., they require long computation times or yield low-quality solutions.
Therefore, real-world scheduling applications often resort to fast,
handcrafted, priority-based dispatching heuristics to find a good initial
solution and then refine it using optimization methods.
This paper proposes a novel end-to-end approach to solving scheduling
problems by means of CP and Reinforcement Learning (RL). In contrast to
previous RL methods, tailored for a given problem by including procedural
simulation algorithms, complex feature engineering, or handcrafted reward
functions, our neural-network architecture and training algorithm merely
require a generic CP encoding of some scheduling problem along with a set of
small instances. Our approach leverages existing CP solvers to train an agent
learning a Priority Dispatching Rule (PDR) that generalizes well to large
instances, even from separate datasets. We evaluate our method on seven JSSP
datasets from the literature, showing its ability to find higher-quality
solutions for very large instances than obtained by static PDRs and by a CP
solver within the same time limit.
|
2306.05760 | Yao Rong | Yao Rong, Guanchu Wang, Qizhang Feng, Ninghao Liu, Zirui Liu,
Enkelejda Kasneci, Xia Hu | Efficient GNN Explanation via Learning Removal-based Attribution | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As Graph Neural Networks (GNNs) have been widely used in real-world
applications, model explanations are required not only by users but also by
legal regulations. However, simultaneously achieving high fidelity and low
computational costs in generating explanations has been a challenge for current
methods. In this work, we propose a framework of GNN explanation named LeArn
Removal-based Attribution (LARA) to address this problem. Specifically, we
introduce removal-based attribution and demonstrate its substantiated link to
interpretability fidelity theoretically and experimentally. The explainer in
LARA learns to generate removal-based attribution which enables providing
explanations with high fidelity. A strategy of subgraph sampling is designed in
LARA to improve the scalability of the training process. In the deployment,
LARA can efficiently generate the explanation through a feed-forward pass. We
benchmark our approach with other state-of-the-art GNN explanation methods on
six datasets. Results highlight the effectiveness of our framework regarding
both efficiency and fidelity. In particular, LARA is 3.5 times faster and
achieves higher fidelity than the state-of-the-art method on the large dataset
ogbn-arxiv (more than 160K nodes and 1M edges), showing its great potential in
real-world applications. Our source code is available at
https://anonymous.4open.science/r/LARA-10D8/README.md.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 08:54:20 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Rong",
"Yao",
""
],
[
"Wang",
"Guanchu",
""
],
[
"Feng",
"Qizhang",
""
],
[
"Liu",
"Ninghao",
""
],
[
"Liu",
"Zirui",
""
],
[
"Kasneci",
"Enkelejda",
""
],
[
"Hu",
"Xia",
""
]
] | TITLE: Efficient GNN Explanation via Learning Removal-based Attribution
ABSTRACT: As Graph Neural Networks (GNNs) have been widely used in real-world
applications, model explanations are required not only by users but also by
legal regulations. However, simultaneously achieving high fidelity and low
computational costs in generating explanations has been a challenge for current
methods. In this work, we propose a framework of GNN explanation named LeArn
Removal-based Attribution (LARA) to address this problem. Specifically, we
introduce removal-based attribution and demonstrate its substantiated link to
interpretability fidelity theoretically and experimentally. The explainer in
LARA learns to generate removal-based attribution which enables providing
explanations with high fidelity. A strategy of subgraph sampling is designed in
LARA to improve the scalability of the training process. In the deployment,
LARA can efficiently generate the explanation through a feed-forward pass. We
benchmark our approach with other state-of-the-art GNN explanation methods on
six datasets. Results highlight the effectiveness of our framework regarding
both efficiency and fidelity. In particular, LARA is 3.5 times faster and
achieves higher fidelity than the state-of-the-art method on the large dataset
ogbn-arxiv (more than 160K nodes and 1M edges), showing its great potential in
real-world applications. Our source code is available at
https://anonymous.4open.science/r/LARA-10D8/README.md.
|
2306.05766 | Sunny Katyara | Sunny Katyara, Mohammad Mujtahid, Court Edmondson | Data-Link: High Fidelity Manufacturing Datasets for Model2Real Transfer
under Industrial Settings | RSS 2023 | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | High-fidelity datasets play a pivotal role in imbuing simulators with
realism, enabling the benchmarking of various state-of-the-art deep inference
models. These models are particularly instrumental in tasks such as semantic
segmentation, classification, and localization. This study showcases the
efficacy of a customized manufacturing dataset comprising 60 classes in the
creation of a high-fidelity digital twin of a robotic manipulation environment.
By leveraging the concept of transfer learning, different 6D pose estimation
models are trained within the simulated environment using domain randomization
and subsequently tested on real-world objects to assess domain adaptation. To
ascertain the effectiveness and realism of the created data-set, pose accuracy
and mean absolute error (MAE) metrics are reported to quantify the model2real
gap.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 09:04:35 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Katyara",
"Sunny",
""
],
[
"Mujtahid",
"Mohammad",
""
],
[
"Edmondson",
"Court",
""
]
] | TITLE: Data-Link: High Fidelity Manufacturing Datasets for Model2Real Transfer
under Industrial Settings
ABSTRACT: High-fidelity datasets play a pivotal role in imbuing simulators with
realism, enabling the benchmarking of various state-of-the-art deep inference
models. These models are particularly instrumental in tasks such as semantic
segmentation, classification, and localization. This study showcases the
efficacy of a customized manufacturing dataset comprising 60 classes in the
creation of a high-fidelity digital twin of a robotic manipulation environment.
By leveraging the concept of transfer learning, different 6D pose estimation
models are trained within the simulated environment using domain randomization
and subsequently tested on real-world objects to assess domain adaptation. To
ascertain the effectiveness and realism of the created data-set, pose accuracy
and mean absolute error (MAE) metrics are reported to quantify the model2real
gap.
|
2306.05776 | Michael K\"olle | Michael K\"olle, Alessandro Giovagnoli, Jonas Stein, Maximilian
Balthasar Mansky, Julian Hager, Tobias Rohe, Robert M\"uller and Claudia
Linnhoff-Popien | Weight Re-Mapping for Variational Quantum Algorithms | null | null | null | null | quant-ph cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Inspired by the remarkable success of artificial neural networks across a
broad spectrum of AI tasks, variational quantum circuits (VQCs) have recently
seen an upsurge in quantum machine learning applications. The promising
outcomes shown by VQCs, such as improved generalization and reduced parameter
training requirements, are attributed to the robust algorithmic capabilities of
quantum computing. However, the current gradient-based training approaches for
VQCs do not adequately accommodate the fact that trainable parameters (or
weights) are typically used as angles in rotational gates. To address this, we
extend the concept of weight re-mapping for VQCs, as introduced by K\"olle et
al. (2023). This approach unambiguously maps the weights to an interval of
length $2\pi$, mirroring data rescaling techniques in conventional machine
learning that have proven to be highly beneficial in numerous scenarios. In our
study, we employ seven distinct weight re-mapping functions to assess their
impact on eight classification datasets, using variational classifiers as a
representative example. Our results indicate that weight re-mapping can enhance
the convergence speed of the VQC. We assess the efficacy of various re-mapping
functions across all datasets and measure their influence on the VQC's average
performance. Our findings indicate that weight re-mapping not only consistently
accelerates the convergence of VQCs, regardless of the specific re-mapping
function employed, but also significantly increases accuracy in certain cases.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 09:42:21 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Kölle",
"Michael",
""
],
[
"Giovagnoli",
"Alessandro",
""
],
[
"Stein",
"Jonas",
""
],
[
"Mansky",
"Maximilian Balthasar",
""
],
[
"Hager",
"Julian",
""
],
[
"Rohe",
"Tobias",
""
],
[
"Müller",
"Robert",
""
],
[
"Linnhoff-Popien",
"Claudia",
""
]
] | TITLE: Weight Re-Mapping for Variational Quantum Algorithms
ABSTRACT: Inspired by the remarkable success of artificial neural networks across a
broad spectrum of AI tasks, variational quantum circuits (VQCs) have recently
seen an upsurge in quantum machine learning applications. The promising
outcomes shown by VQCs, such as improved generalization and reduced parameter
training requirements, are attributed to the robust algorithmic capabilities of
quantum computing. However, the current gradient-based training approaches for
VQCs do not adequately accommodate the fact that trainable parameters (or
weights) are typically used as angles in rotational gates. To address this, we
extend the concept of weight re-mapping for VQCs, as introduced by K\"olle et
al. (2023). This approach unambiguously maps the weights to an interval of
length $2\pi$, mirroring data rescaling techniques in conventional machine
learning that have proven to be highly beneficial in numerous scenarios. In our
study, we employ seven distinct weight re-mapping functions to assess their
impact on eight classification datasets, using variational classifiers as a
representative example. Our results indicate that weight re-mapping can enhance
the convergence speed of the VQC. We assess the efficacy of various re-mapping
functions across all datasets and measure their influence on the VQC's average
performance. Our findings indicate that weight re-mapping not only consistently
accelerates the convergence of VQCs, regardless of the specific re-mapping
function employed, but also significantly increases accuracy in certain cases.
|
2306.05784 | Aneeqa Abrar | Aneeqa Abrar, Hamza Iqbal | Quantitative Ink Analysis: Estimating the Number of Inks in Documents
through Hyperspectral Imaging | null | null | null | null | cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | In the field of document forensics, ink analysis plays a crucial role in
determining the authenticity of legal and historic documents and detecting
forgery. Visual examination alone is insufficient for distinguishing visually
similar inks, necessitating the use of advanced scientific techniques. This
paper proposes an ink analysis technique based on hyperspectral imaging, which
enables the examination of documents in hundreds of narrowly spaced spectral
bands, revealing hidden details. The main objective of this study is to
identify the number of distinct inks used in a document. Three clustering
algorithms, namely k-means, Agglomerative, and c-means, are employed to
estimate the number of inks present. The methodology involves data extraction,
ink pixel segmentation, and ink number determination. The results demonstrate
the effectiveness of the proposed technique in identifying ink clusters and
distinguishing between different inks. The analysis of a hyperspectral cube
dataset reveals variations in spectral reflectance across different bands and
distinct spectral responses among the 12 lines, indicating the presence of
multiple inks. The clustering algorithms successfully identify ink clusters,
with k-means clustering showing superior classification performance. These
findings contribute to the development of reliable methodologies for ink
analysis using hyperspectral imaging, enhancing the
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 09:55:20 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Abrar",
"Aneeqa",
""
],
[
"Iqbal",
"Hamza",
""
]
] | TITLE: Quantitative Ink Analysis: Estimating the Number of Inks in Documents
through Hyperspectral Imaging
ABSTRACT: In the field of document forensics, ink analysis plays a crucial role in
determining the authenticity of legal and historic documents and detecting
forgery. Visual examination alone is insufficient for distinguishing visually
similar inks, necessitating the use of advanced scientific techniques. This
paper proposes an ink analysis technique based on hyperspectral imaging, which
enables the examination of documents in hundreds of narrowly spaced spectral
bands, revealing hidden details. The main objective of this study is to
identify the number of distinct inks used in a document. Three clustering
algorithms, namely k-means, Agglomerative, and c-means, are employed to
estimate the number of inks present. The methodology involves data extraction,
ink pixel segmentation, and ink number determination. The results demonstrate
the effectiveness of the proposed technique in identifying ink clusters and
distinguishing between different inks. The analysis of a hyperspectral cube
dataset reveals variations in spectral reflectance across different bands and
distinct spectral responses among the 12 lines, indicating the presence of
multiple inks. The clustering algorithms successfully identify ink clusters,
with k-means clustering showing superior classification performance. These
findings contribute to the development of reliable methodologies for ink
analysis using hyperspectral imaging, enhancing the
|
2306.05801 | Andrea Apicella | Andrea Apicella, Luca Di Lorenzo, Francesco Isgr\`o, Andrea Pollastro,
Roberto Prevete | Strategies to exploit XAI to improve classification systems | This work has been accepted to be presented to The 1st World
Conference on eXplainable Artificial Intelligence (xAI 2023), July 26-28,
2023 - Lisboa, Portugal | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Explainable Artificial Intelligence (XAI) aims to provide insights into the
decision-making process of AI models, allowing users to understand their
results beyond their decisions. A significant goal of XAI is to improve the
performance of AI models by providing explanations for their decision-making
processes. However, most XAI literature focuses on how to explain an AI system,
while less attention has been given to how XAI methods can be exploited to
improve an AI system. In this work, a set of well-known XAI methods typically
used with Machine Learning (ML) classification tasks are investigated to verify
if they can be exploited, not just to provide explanations but also to improve
the performance of the model itself. To this aim, two strategies to use the
explanation to improve a classification system are reported and empirically
evaluated on three datasets: Fashion-MNIST, CIFAR10, and STL10. Results suggest
that explanations built by Integrated Gradients highlight input features that
can be effectively used to improve classification performance.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 10:38:26 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Apicella",
"Andrea",
""
],
[
"Di Lorenzo",
"Luca",
""
],
[
"Isgrò",
"Francesco",
""
],
[
"Pollastro",
"Andrea",
""
],
[
"Prevete",
"Roberto",
""
]
] | TITLE: Strategies to exploit XAI to improve classification systems
ABSTRACT: Explainable Artificial Intelligence (XAI) aims to provide insights into the
decision-making process of AI models, allowing users to understand their
results beyond their decisions. A significant goal of XAI is to improve the
performance of AI models by providing explanations for their decision-making
processes. However, most XAI literature focuses on how to explain an AI system,
while less attention has been given to how XAI methods can be exploited to
improve an AI system. In this work, a set of well-known XAI methods typically
used with Machine Learning (ML) classification tasks are investigated to verify
if they can be exploited, not just to provide explanations but also to improve
the performance of the model itself. To this aim, two strategies to use the
explanation to improve a classification system are reported and empirically
evaluated on three datasets: Fashion-MNIST, CIFAR10, and STL10. Results suggest
that explanations built by Integrated Gradients highlight input features that
can be effectively used to improve classification performance.
|
2306.05808 | Maarten Buyl | Maarten Buyl, Paul Missault and Pierre-Antoine Sondag | RankFormer: Listwise Learning-to-Rank Using Listwide Labels | Accepted at KDD 2023 | null | 10.1145/3580305.3599892 | null | cs.IR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Web applications where users are presented with a limited selection of items
have long employed ranking models to put the most relevant results first. Any
feedback received from users is typically assumed to reflect a relative
judgement on the utility of items, e.g. a user clicking on an item only implies
it is better than items not clicked in the same ranked list. Hence, the
objectives optimized in Learning-to-Rank (LTR) tend to be pairwise or listwise.
Yet, by only viewing feedback as relative, we neglect the user's absolute
feedback on the list's overall quality, e.g. when no items in the selection are
clicked. We thus reconsider the standard LTR paradigm and argue the benefits of
learning from this listwide signal. To this end, we propose the RankFormer as
an architecture that, with a Transformer at its core, can jointly optimize a
novel listwide assessment objective and a traditional listwise LTR objective.
We simulate implicit feedback on public datasets and observe that the
RankFormer succeeds in benefitting from listwide signals. Additionally, we
conduct experiments in e-commerce on Amazon Search data and find the RankFormer
to be superior to all baselines offline. An online experiment shows that
knowledge distillation can be used to find immediate practical use for the
RankFormer.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 10:47:06 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Buyl",
"Maarten",
""
],
[
"Missault",
"Paul",
""
],
[
"Sondag",
"Pierre-Antoine",
""
]
] | TITLE: RankFormer: Listwise Learning-to-Rank Using Listwide Labels
ABSTRACT: Web applications where users are presented with a limited selection of items
have long employed ranking models to put the most relevant results first. Any
feedback received from users is typically assumed to reflect a relative
judgement on the utility of items, e.g. a user clicking on an item only implies
it is better than items not clicked in the same ranked list. Hence, the
objectives optimized in Learning-to-Rank (LTR) tend to be pairwise or listwise.
Yet, by only viewing feedback as relative, we neglect the user's absolute
feedback on the list's overall quality, e.g. when no items in the selection are
clicked. We thus reconsider the standard LTR paradigm and argue the benefits of
learning from this listwide signal. To this end, we propose the RankFormer as
an architecture that, with a Transformer at its core, can jointly optimize a
novel listwide assessment objective and a traditional listwise LTR objective.
We simulate implicit feedback on public datasets and observe that the
RankFormer succeeds in benefitting from listwide signals. Additionally, we
conduct experiments in e-commerce on Amazon Search data and find the RankFormer
to be superior to all baselines offline. An online experiment shows that
knowledge distillation can be used to find immediate practical use for the
RankFormer.
|
2306.05815 | Francesco Tonin | Francesco Tonin, Alex Lambert, Panagiotis Patrinos, Johan A. K.
Suykens | Extending Kernel PCA through Dualization: Sparsity, Robustness and Fast
Algorithms | 15 pages, ICML 2023 | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The goal of this paper is to revisit Kernel Principal Component Analysis
(KPCA) through dualization of a difference of convex functions. This allows to
naturally extend KPCA to multiple objective functions and leads to efficient
gradient-based algorithms avoiding the expensive SVD of the Gram matrix.
Particularly, we consider objective functions that can be written as Moreau
envelopes, demonstrating how to promote robustness and sparsity within the same
framework. The proposed method is evaluated on synthetic and real-world
benchmarks, showing significant speedup in KPCA training time as well as
highlighting the benefits in terms of robustness and sparsity.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 11:27:35 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Tonin",
"Francesco",
""
],
[
"Lambert",
"Alex",
""
],
[
"Patrinos",
"Panagiotis",
""
],
[
"Suykens",
"Johan A. K.",
""
]
] | TITLE: Extending Kernel PCA through Dualization: Sparsity, Robustness and Fast
Algorithms
ABSTRACT: The goal of this paper is to revisit Kernel Principal Component Analysis
(KPCA) through dualization of a difference of convex functions. This allows to
naturally extend KPCA to multiple objective functions and leads to efficient
gradient-based algorithms avoiding the expensive SVD of the Gram matrix.
Particularly, we consider objective functions that can be written as Moreau
envelopes, demonstrating how to promote robustness and sparsity within the same
framework. The proposed method is evaluated on synthetic and real-world
benchmarks, showing significant speedup in KPCA training time as well as
highlighting the benefits in terms of robustness and sparsity.
|
2306.05844 | Dustin Aganian | Dustin Aganian, Mona K\"ohler, Sebastian Baake, Markus Eisenbach, and
Horst-Michael Gross | How Object Information Improves Skeleton-based Human Action Recognition
in Assembly Tasks | IEEE International Joint Conference on Neural Networks (IJCNN) 2023 | null | null | null | cs.CV cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As the use of collaborative robots (cobots) in industrial manufacturing
continues to grow, human action recognition for effective human-robot
collaboration becomes increasingly important. This ability is crucial for
cobots to act autonomously and assist in assembly tasks. Recently,
skeleton-based approaches are often used as they tend to generalize better to
different people and environments. However, when processing skeletons alone,
information about the objects a human interacts with is lost. Therefore, we
present a novel approach of integrating object information into skeleton-based
action recognition. We enhance two state-of-the-art methods by treating object
centers as further skeleton joints. Our experiments on the assembly dataset
IKEA ASM show that our approach improves the performance of these
state-of-the-art methods to a large extent when combining skeleton joints with
objects predicted by a state-of-the-art instance segmentation model. Our
research sheds light on the benefits of combining skeleton joints with object
information for human action recognition in assembly tasks. We analyze the
effect of the object detector on the combination for action classification and
discuss the important factors that must be taken into account.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 12:18:14 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Aganian",
"Dustin",
""
],
[
"Köhler",
"Mona",
""
],
[
"Baake",
"Sebastian",
""
],
[
"Eisenbach",
"Markus",
""
],
[
"Gross",
"Horst-Michael",
""
]
] | TITLE: How Object Information Improves Skeleton-based Human Action Recognition
in Assembly Tasks
ABSTRACT: As the use of collaborative robots (cobots) in industrial manufacturing
continues to grow, human action recognition for effective human-robot
collaboration becomes increasingly important. This ability is crucial for
cobots to act autonomously and assist in assembly tasks. Recently,
skeleton-based approaches are often used as they tend to generalize better to
different people and environments. However, when processing skeletons alone,
information about the objects a human interacts with is lost. Therefore, we
present a novel approach of integrating object information into skeleton-based
action recognition. We enhance two state-of-the-art methods by treating object
centers as further skeleton joints. Our experiments on the assembly dataset
IKEA ASM show that our approach improves the performance of these
state-of-the-art methods to a large extent when combining skeleton joints with
objects predicted by a state-of-the-art instance segmentation model. Our
research sheds light on the benefits of combining skeleton joints with object
information for human action recognition in assembly tasks. We analyze the
effect of the object detector on the combination for action classification and
discuss the important factors that must be taken into account.
|
2306.05854 | Amalee Wilson | Amalee Wilson, Andres Noetzli, Andrew Reynolds, Byron Cook, Cesare
Tinelli, and Clark Barrett | Partitioning Strategies for Distributed SMT Solving | Submitted to FMCAD 2023 | null | null | null | cs.DC cs.LO | http://creativecommons.org/licenses/by/4.0/ | For many users of Satisfiability Modulo Theories (SMT) solvers, the solver's
performance is the main bottleneck in their application. One promising approach
for improving performance is to leverage the increasing availability of
parallel and cloud computing. However, despite many efforts, the best parallel
approach to date consists of running a portfolio of solvers, meaning that
performance is still limited by the best possible sequential performance. In
this paper, we revisit divide-and-conquer approaches to parallel SMT, in which
a challenging problem is partitioned into several subproblems. We introduce
several new partitioning strategies and evaluate their performance, both alone
as well as within portfolios, on a large set of difficult SMT benchmarks. We
show that hybrid portfolios that include our new strategies can significantly
outperform traditional portfolios for parallel SMT.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 16:53:59 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wilson",
"Amalee",
""
],
[
"Noetzli",
"Andres",
""
],
[
"Reynolds",
"Andrew",
""
],
[
"Cook",
"Byron",
""
],
[
"Tinelli",
"Cesare",
""
],
[
"Barrett",
"Clark",
""
]
] | TITLE: Partitioning Strategies for Distributed SMT Solving
ABSTRACT: For many users of Satisfiability Modulo Theories (SMT) solvers, the solver's
performance is the main bottleneck in their application. One promising approach
for improving performance is to leverage the increasing availability of
parallel and cloud computing. However, despite many efforts, the best parallel
approach to date consists of running a portfolio of solvers, meaning that
performance is still limited by the best possible sequential performance. In
this paper, we revisit divide-and-conquer approaches to parallel SMT, in which
a challenging problem is partitioned into several subproblems. We introduce
several new partitioning strategies and evaluate their performance, both alone
as well as within portfolios, on a large set of difficult SMT benchmarks. We
show that hybrid portfolios that include our new strategies can significantly
outperform traditional portfolios for parallel SMT.
|
2306.05861 | Junyu Wang | Junyu Wang | Efficient Encoder-Decoder and Dual-Path Conformer for Comprehensive
Feature Learning in Speech Enhancement | Accepted at Interspeech2023 | null | null | null | eess.AS cs.CL cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current speech enhancement (SE) research has largely neglected channel
attention and spatial attention, and encoder-decoder architecture-based
networks have not adequately considered how to provide efficient inputs to the
intermediate enhancement layer. To address these issues, this paper proposes a
time-frequency (T-F) domain SE network (DPCFCS-Net) that incorporates improved
densely connected blocks, dual-path modules, convolution-augmented transformers
(conformers), channel attention, and spatial attention. Compared with previous
models, our proposed model has a more efficient encoder-decoder and can learn
comprehensive features. Experimental results on the VCTK+DEMAND dataset
demonstrate that our method outperforms existing techniques in SE performance.
Furthermore, the improved densely connected block and two dimensions attention
module developed in this work are highly adaptable and easily integrated into
existing networks.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 12:52:01 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wang",
"Junyu",
""
]
] | TITLE: Efficient Encoder-Decoder and Dual-Path Conformer for Comprehensive
Feature Learning in Speech Enhancement
ABSTRACT: Current speech enhancement (SE) research has largely neglected channel
attention and spatial attention, and encoder-decoder architecture-based
networks have not adequately considered how to provide efficient inputs to the
intermediate enhancement layer. To address these issues, this paper proposes a
time-frequency (T-F) domain SE network (DPCFCS-Net) that incorporates improved
densely connected blocks, dual-path modules, convolution-augmented transformers
(conformers), channel attention, and spatial attention. Compared with previous
models, our proposed model has a more efficient encoder-decoder and can learn
comprehensive features. Experimental results on the VCTK+DEMAND dataset
demonstrate that our method outperforms existing techniques in SE performance.
Furthermore, the improved densely connected block and two dimensions attention
module developed in this work are highly adaptable and easily integrated into
existing networks.
|
2306.05871 | Wissam Antoun | Wissam Antoun, Virginie Mouilleron, Beno\^it Sagot, Djam\'e Seddah | Towards a Robust Detection of Language Model Generated Text: Is ChatGPT
that Easy to Detect? | Accepted to TALN 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Recent advances in natural language processing (NLP) have led to the
development of large language models (LLMs) such as ChatGPT. This paper
proposes a methodology for developing and evaluating ChatGPT detectors for
French text, with a focus on investigating their robustness on out-of-domain
data and against common attack schemes. The proposed method involves
translating an English dataset into French and training a classifier on the
translated data. Results show that the detectors can effectively detect
ChatGPT-generated text, with a degree of robustness against basic attack
techniques in in-domain settings. However, vulnerabilities are evident in
out-of-domain contexts, highlighting the challenge of detecting adversarial
text. The study emphasizes caution when applying in-domain testing results to a
wider variety of content. We provide our translated datasets and models as
open-source resources. https://gitlab.inria.fr/wantoun/robust-chatgpt-detection
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 13:03:53 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Antoun",
"Wissam",
""
],
[
"Mouilleron",
"Virginie",
""
],
[
"Sagot",
"Benoît",
""
],
[
"Seddah",
"Djamé",
""
]
] | TITLE: Towards a Robust Detection of Language Model Generated Text: Is ChatGPT
that Easy to Detect?
ABSTRACT: Recent advances in natural language processing (NLP) have led to the
development of large language models (LLMs) such as ChatGPT. This paper
proposes a methodology for developing and evaluating ChatGPT detectors for
French text, with a focus on investigating their robustness on out-of-domain
data and against common attack schemes. The proposed method involves
translating an English dataset into French and training a classifier on the
translated data. Results show that the detectors can effectively detect
ChatGPT-generated text, with a degree of robustness against basic attack
techniques in in-domain settings. However, vulnerabilities are evident in
out-of-domain contexts, highlighting the challenge of detecting adversarial
text. The study emphasizes caution when applying in-domain testing results to a
wider variety of content. We provide our translated datasets and models as
open-source resources. https://gitlab.inria.fr/wantoun/robust-chatgpt-detection
|
2306.05895 | Tomas Cerny Ph.D. | Sheldon Smith, Ethan Robinson, Timmy Frederiksen, Trae Stevens, Tomas
Cerny, Miroslav Bures, Davide Taibi | Benchmarks for End-to-End Microservices Testing | 7 pages | IEEE SOSE 2023 | null | null | cs.SE cs.DC | http://creativecommons.org/licenses/by/4.0/ | Testing microservice systems involves a large amount of planning and
problem-solving. The difficulty of testing microservice systems increases as
the size and structure of such systems become more complex. To help the
microservice community and simplify experiments with testing and traffic
simulation, we created a test benchmark containing full functional testing
coverage for two well-established open-source microservice systems. Through our
benchmark design, we aimed to demonstrate ways to overcome certain challenges
and find effective strategies when testing microservices. In addition, to
demonstrate our benchmark use, we conducted a case study to identify the best
approaches to take to validate a full coverage of tests using
service-dependency graph discovery and business process discovery using
tracing.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 13:42:53 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Smith",
"Sheldon",
""
],
[
"Robinson",
"Ethan",
""
],
[
"Frederiksen",
"Timmy",
""
],
[
"Stevens",
"Trae",
""
],
[
"Cerny",
"Tomas",
""
],
[
"Bures",
"Miroslav",
""
],
[
"Taibi",
"Davide",
""
]
] | TITLE: Benchmarks for End-to-End Microservices Testing
ABSTRACT: Testing microservice systems involves a large amount of planning and
problem-solving. The difficulty of testing microservice systems increases as
the size and structure of such systems become more complex. To help the
microservice community and simplify experiments with testing and traffic
simulation, we created a test benchmark containing full functional testing
coverage for two well-established open-source microservice systems. Through our
benchmark design, we aimed to demonstrate ways to overcome certain challenges
and find effective strategies when testing microservices. In addition, to
demonstrate our benchmark use, we conducted a case study to identify the best
approaches to take to validate a full coverage of tests using
service-dependency graph discovery and business process discovery using
tracing.
|
2306.05915 | Neil Scheidwasser | Zihan Wu, Neil Scheidwasser-Clow, Karl El Hajal, Milos Cernak | Speaker Embeddings as Individuality Proxy for Voice Stress Detection | 5 pages, 2 figures. Accepted at Interspeech 2023 | null | null | null | eess.AS cs.LG | http://creativecommons.org/licenses/by/4.0/ | Since the mental states of the speaker modulate speech, stress introduced by
cognitive or physical loads could be detected in the voice. The existing voice
stress detection benchmark has shown that the audio embeddings extracted from
the Hybrid BYOL-S self-supervised model perform well. However, the benchmark
only evaluates performance separately on each dataset, but does not evaluate
performance across the different types of stress and different languages.
Moreover, previous studies found strong individual differences in stress
susceptibility. This paper presents the design and development of voice stress
detection, trained on more than 100 speakers from 9 language groups and five
different types of stress. We address individual variabilities in voice stress
analysis by adding speaker embeddings to the hybrid BYOL-S features. The
proposed method significantly improves voice stress detection performance with
an input audio length of only 3-5 seconds.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 14:11:07 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Wu",
"Zihan",
""
],
[
"Scheidwasser-Clow",
"Neil",
""
],
[
"Hajal",
"Karl El",
""
],
[
"Cernak",
"Milos",
""
]
] | TITLE: Speaker Embeddings as Individuality Proxy for Voice Stress Detection
ABSTRACT: Since the mental states of the speaker modulate speech, stress introduced by
cognitive or physical loads could be detected in the voice. The existing voice
stress detection benchmark has shown that the audio embeddings extracted from
the Hybrid BYOL-S self-supervised model perform well. However, the benchmark
only evaluates performance separately on each dataset, but does not evaluate
performance across the different types of stress and different languages.
Moreover, previous studies found strong individual differences in stress
susceptibility. This paper presents the design and development of voice stress
detection, trained on more than 100 speakers from 9 language groups and five
different types of stress. We address individual variabilities in voice stress
analysis by adding speaker embeddings to the hybrid BYOL-S features. The
proposed method significantly improves voice stress detection performance with
an input audio length of only 3-5 seconds.
|
2306.05955 | Johannes Lutzeyer | Gaspard Michel, Giannis Nikolentzos, Johannes Lutzeyer, Michalis
Vazirgiannis | Path Neural Networks: Expressive and Accurate Graph Neural Networks | Accepted at ICML 2023 | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph neural networks (GNNs) have recently become the standard approach for
learning with graph-structured data. Prior work has shed light into their
potential, but also their limitations. Unfortunately, it was shown that
standard GNNs are limited in their expressive power. These models are no more
powerful than the 1-dimensional Weisfeiler-Leman (1-WL) algorithm in terms of
distinguishing non-isomorphic graphs. In this paper, we propose Path Neural
Networks (PathNNs), a model that updates node representations by aggregating
paths emanating from nodes. We derive three different variants of the PathNN
model that aggregate single shortest paths, all shortest paths and all simple
paths of length up to K. We prove that two of these variants are strictly more
powerful than the 1-WL algorithm, and we experimentally validate our
theoretical results. We find that PathNNs can distinguish pairs of
non-isomorphic graphs that are indistinguishable by 1-WL, while our most
expressive PathNN variant can even distinguish between 3-WL indistinguishable
graphs. The different PathNN variants are also evaluated on graph
classification and graph regression datasets, where in most cases, they
outperform the baseline methods.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 15:11:49 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Michel",
"Gaspard",
""
],
[
"Nikolentzos",
"Giannis",
""
],
[
"Lutzeyer",
"Johannes",
""
],
[
"Vazirgiannis",
"Michalis",
""
]
] | TITLE: Path Neural Networks: Expressive and Accurate Graph Neural Networks
ABSTRACT: Graph neural networks (GNNs) have recently become the standard approach for
learning with graph-structured data. Prior work has shed light into their
potential, but also their limitations. Unfortunately, it was shown that
standard GNNs are limited in their expressive power. These models are no more
powerful than the 1-dimensional Weisfeiler-Leman (1-WL) algorithm in terms of
distinguishing non-isomorphic graphs. In this paper, we propose Path Neural
Networks (PathNNs), a model that updates node representations by aggregating
paths emanating from nodes. We derive three different variants of the PathNN
model that aggregate single shortest paths, all shortest paths and all simple
paths of length up to K. We prove that two of these variants are strictly more
powerful than the 1-WL algorithm, and we experimentally validate our
theoretical results. We find that PathNNs can distinguish pairs of
non-isomorphic graphs that are indistinguishable by 1-WL, while our most
expressive PathNN variant can even distinguish between 3-WL indistinguishable
graphs. The different PathNN variants are also evaluated on graph
classification and graph regression datasets, where in most cases, they
outperform the baseline methods.
|
2306.05985 | Luka Dragar | Luka Dragar, Peter Peer, Vitomir \v{S}truc, Borut Batagelj | Beyond Detection: Visual Realism Assessment of Deepfakes | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the era of rapid digitalization and artificial intelligence advancements,
the development of DeepFake technology has posed significant security and
privacy concerns. This paper presents an effective measure to assess the visual
realism of DeepFake videos. We utilize an ensemble of two Convolutional Neural
Network (CNN) models: Eva and ConvNext. These models have been trained on the
DeepFake Game Competition (DFGC) 2022 dataset and aim to predict Mean Opinion
Scores (MOS) from DeepFake videos based on features extracted from sequences of
frames. Our method secured the third place in the recent DFGC on Visual Realism
Assessment held in conjunction with the 2023 International Joint Conference on
Biometrics (IJCB 2023). We provide an over\-view of the models, data
preprocessing, and training procedures. We also report the performance of our
models against the competition's baseline model and discuss the implications of
our findings.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 15:53:01 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Dragar",
"Luka",
""
],
[
"Peer",
"Peter",
""
],
[
"Štruc",
"Vitomir",
""
],
[
"Batagelj",
"Borut",
""
]
] | TITLE: Beyond Detection: Visual Realism Assessment of Deepfakes
ABSTRACT: In the era of rapid digitalization and artificial intelligence advancements,
the development of DeepFake technology has posed significant security and
privacy concerns. This paper presents an effective measure to assess the visual
realism of DeepFake videos. We utilize an ensemble of two Convolutional Neural
Network (CNN) models: Eva and ConvNext. These models have been trained on the
DeepFake Game Competition (DFGC) 2022 dataset and aim to predict Mean Opinion
Scores (MOS) from DeepFake videos based on features extracted from sequences of
frames. Our method secured the third place in the recent DFGC on Visual Realism
Assessment held in conjunction with the 2023 International Joint Conference on
Biometrics (IJCB 2023). We provide an over\-view of the models, data
preprocessing, and training procedures. We also report the performance of our
models against the competition's baseline model and discuss the implications of
our findings.
|
2306.05998 | Xiaotong Cheng | Xiaotong Cheng, Setareh Maghsudi | Distributed Consensus Algorithm for Decision-Making in Multi-agent
Multi-armed Bandit | null | null | null | null | cs.LG cs.MA stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study a structured multi-agent multi-armed bandit (MAMAB) problem in a
dynamic environment. A graph reflects the information-sharing structure among
agents, and the arms' reward distributions are piecewise-stationary with
several unknown change points. The agents face the identical
piecewise-stationary MAB problem. The goal is to develop a decision-making
policy for the agents that minimizes the regret, which is the expected total
loss of not playing the optimal arm at each time step. Our proposed solution,
Restarted Bayesian Online Change Point Detection in Cooperative Upper
Confidence Bound Algorithm (RBO-Coop-UCB), involves an efficient multi-agent
UCB algorithm as its core enhanced with a Bayesian change point detector. We
also develop a simple restart decision cooperation that improves
decision-making. Theoretically, we establish that the expected group regret of
RBO-Coop-UCB is upper bounded by $\mathcal{O}(KNM\log T + K\sqrt{MT\log T})$,
where K is the number of agents, M is the number of arms, and T is the number
of time steps. Numerical experiments on synthetic and real-world datasets
demonstrate that our proposed method outperforms the state-of-the-art
algorithms.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 16:10:26 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Cheng",
"Xiaotong",
""
],
[
"Maghsudi",
"Setareh",
""
]
] | TITLE: Distributed Consensus Algorithm for Decision-Making in Multi-agent
Multi-armed Bandit
ABSTRACT: We study a structured multi-agent multi-armed bandit (MAMAB) problem in a
dynamic environment. A graph reflects the information-sharing structure among
agents, and the arms' reward distributions are piecewise-stationary with
several unknown change points. The agents face the identical
piecewise-stationary MAB problem. The goal is to develop a decision-making
policy for the agents that minimizes the regret, which is the expected total
loss of not playing the optimal arm at each time step. Our proposed solution,
Restarted Bayesian Online Change Point Detection in Cooperative Upper
Confidence Bound Algorithm (RBO-Coop-UCB), involves an efficient multi-agent
UCB algorithm as its core enhanced with a Bayesian change point detector. We
also develop a simple restart decision cooperation that improves
decision-making. Theoretically, we establish that the expected group regret of
RBO-Coop-UCB is upper bounded by $\mathcal{O}(KNM\log T + K\sqrt{MT\log T})$,
where K is the number of agents, M is the number of arms, and T is the number
of time steps. Numerical experiments on synthetic and real-world datasets
demonstrate that our proposed method outperforms the state-of-the-art
algorithms.
|
2306.06007 | Sepand Kashani | Sepand Kashani, Joan Ru\'e Queralt, Adrian Jarret, Matthieu Simeoni | HVOX: Scalable Interferometric Synthesis and Analysis of Spherical Sky
Maps | null | null | null | null | cs.CE astro-ph.IM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Analysis and synthesis are key steps of the radio-interferometric imaging
process, serving as a bridge between visibility and sky domains. They can be
expressed as partial Fourier transforms involving a large number of non-uniform
frequencies and spherically-constrained spatial coordinates. Due to the data
non-uniformity, these partial Fourier transforms are computationally expensive
and represent a serious bottleneck in the image reconstruction process. The
W-gridding algorithm achieves log-linear complexity for both steps by applying
a series of 2D non-uniform FFTs (NUFFT) to the data sliced along the so-called
$w$ frequency coordinate. A major drawback of this method however is its
restriction to direction-cosine meshes, which are fundamentally ill-suited for
large field of views. This paper introduces the HVOX gridder, a novel algorithm
for analysis/synthesis based on a 3D-NUFFT. Unlike W-gridding, the latter is
compatible with arbitrary spherical meshes such as the popular HEALPix scheme
for spherical data processing. The 3D-NUFFT allows one to optimally select the
size of the inner FFTs, in particular the number of W-planes. This results in a
better performing and auto-tuned algorithm, with controlled accuracy guarantees
backed by strong results from approximation theory. To cope with the
challenging scale of next-generation radio telescopes, we propose moreover a
chunked evaluation strategy: by partitioning the visibility and sky domains,
the 3D-NUFFT is decomposed into sub-problems which execute in parallel, while
simultaneously cutting memory requirements. Our benchmarking results
demonstrate the scalability of HVOX for both SKA and LOFAR, considering
state-of-the-art challenging imaging setups. HVOX is moreover computationally
competitive with W-gridder, despite the absence of domain-specific
optimizations in our implementation.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 16:22:32 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Kashani",
"Sepand",
""
],
[
"Queralt",
"Joan Rué",
""
],
[
"Jarret",
"Adrian",
""
],
[
"Simeoni",
"Matthieu",
""
]
] | TITLE: HVOX: Scalable Interferometric Synthesis and Analysis of Spherical Sky
Maps
ABSTRACT: Analysis and synthesis are key steps of the radio-interferometric imaging
process, serving as a bridge between visibility and sky domains. They can be
expressed as partial Fourier transforms involving a large number of non-uniform
frequencies and spherically-constrained spatial coordinates. Due to the data
non-uniformity, these partial Fourier transforms are computationally expensive
and represent a serious bottleneck in the image reconstruction process. The
W-gridding algorithm achieves log-linear complexity for both steps by applying
a series of 2D non-uniform FFTs (NUFFT) to the data sliced along the so-called
$w$ frequency coordinate. A major drawback of this method however is its
restriction to direction-cosine meshes, which are fundamentally ill-suited for
large field of views. This paper introduces the HVOX gridder, a novel algorithm
for analysis/synthesis based on a 3D-NUFFT. Unlike W-gridding, the latter is
compatible with arbitrary spherical meshes such as the popular HEALPix scheme
for spherical data processing. The 3D-NUFFT allows one to optimally select the
size of the inner FFTs, in particular the number of W-planes. This results in a
better performing and auto-tuned algorithm, with controlled accuracy guarantees
backed by strong results from approximation theory. To cope with the
challenging scale of next-generation radio telescopes, we propose moreover a
chunked evaluation strategy: by partitioning the visibility and sky domains,
the 3D-NUFFT is decomposed into sub-problems which execute in parallel, while
simultaneously cutting memory requirements. Our benchmarking results
demonstrate the scalability of HVOX for both SKA and LOFAR, considering
state-of-the-art challenging imaging setups. HVOX is moreover computationally
competitive with W-gridder, despite the absence of domain-specific
optimizations in our implementation.
|
2306.06029 | Rodrigo Agerri | Rodrigo Agerri, I\~nigo Alonso, Aitziber Atutxa, Ander Berrondo,
Ainara Estarrona, Iker Garcia-Ferrero, Iakes Goenaga, Koldo Gojenola, Maite
Oronoz, Igor Perez-Tejedor, German Rigau and Anar Yeginbergenova | HiTZ@Antidote: Argumentation-driven Explainable Artificial Intelligence
for Digital Medicine | To appear: In SEPLN 2023: 39th International Conference of the
Spanish Society for Natural Language Processing | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Providing high quality explanations for AI predictions based on machine
learning is a challenging and complex task. To work well it requires, among
other factors: selecting a proper level of generality/specificity of the
explanation; considering assumptions about the familiarity of the explanation
beneficiary with the AI task under consideration; referring to specific
elements that have contributed to the decision; making use of additional
knowledge (e.g. expert evidence) which might not be part of the prediction
process; and providing evidence supporting negative hypothesis. Finally, the
system needs to formulate the explanation in a clearly interpretable, and
possibly convincing, way. Given these considerations, ANTIDOTE fosters an
integrated vision of explainable AI, where low-level characteristics of the
deep learning process are combined with higher level schemes proper of the
human argumentation capacity. ANTIDOTE will exploit cross-disciplinary
competences in deep learning and argumentation to support a broader and
innovative view of explainable AI, where the need for high-quality explanations
for clinical cases deliberation is critical. As a first result of the project,
we publish the Antidote CasiMedicos dataset to facilitate research on
explainable AI in general, and argumentation in the medical domain in
particular.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 16:50:02 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Agerri",
"Rodrigo",
""
],
[
"Alonso",
"Iñigo",
""
],
[
"Atutxa",
"Aitziber",
""
],
[
"Berrondo",
"Ander",
""
],
[
"Estarrona",
"Ainara",
""
],
[
"Garcia-Ferrero",
"Iker",
""
],
[
"Goenaga",
"Iakes",
""
],
[
"Gojenola",
"Koldo",
""
],
[
"Oronoz",
"Maite",
""
],
[
"Perez-Tejedor",
"Igor",
""
],
[
"Rigau",
"German",
""
],
[
"Yeginbergenova",
"Anar",
""
]
] | TITLE: HiTZ@Antidote: Argumentation-driven Explainable Artificial Intelligence
for Digital Medicine
ABSTRACT: Providing high quality explanations for AI predictions based on machine
learning is a challenging and complex task. To work well it requires, among
other factors: selecting a proper level of generality/specificity of the
explanation; considering assumptions about the familiarity of the explanation
beneficiary with the AI task under consideration; referring to specific
elements that have contributed to the decision; making use of additional
knowledge (e.g. expert evidence) which might not be part of the prediction
process; and providing evidence supporting negative hypothesis. Finally, the
system needs to formulate the explanation in a clearly interpretable, and
possibly convincing, way. Given these considerations, ANTIDOTE fosters an
integrated vision of explainable AI, where low-level characteristics of the
deep learning process are combined with higher level schemes proper of the
human argumentation capacity. ANTIDOTE will exploit cross-disciplinary
competences in deep learning and argumentation to support a broader and
innovative view of explainable AI, where the need for high-quality explanations
for clinical cases deliberation is critical. As a first result of the project,
we publish the Antidote CasiMedicos dataset to facilitate research on
explainable AI in general, and argumentation in the medical domain in
particular.
|
2306.06061 | Phomolo Teffo | Teffo Phomolo Nicrocia, Owolawi Pius Adewale, Pholo Moanda Diana | clustering an african hairstyle dataset using pca and k-means | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | The adoption of digital transformation was not expressed in building an
African face shape classifier. In this paper, an approach is presented that
uses k-means to classify African women images. African women rely on beauty
standards recommendations, personal preference, or the newest trends in
hairstyles to decide on the appropriate hairstyle for them. In this paper, an
approach is presented that uses K-means clustering to classify African women's
images. In order to identify potential facial clusters, Haarcascade is used for
feature-based training, and K-means clustering is applied for image
classification.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 14:13:29 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Nicrocia",
"Teffo Phomolo",
""
],
[
"Adewale",
"Owolawi Pius",
""
],
[
"Diana",
"Pholo Moanda",
""
]
] | TITLE: clustering an african hairstyle dataset using pca and k-means
ABSTRACT: The adoption of digital transformation was not expressed in building an
African face shape classifier. In this paper, an approach is presented that
uses k-means to classify African women images. African women rely on beauty
standards recommendations, personal preference, or the newest trends in
hairstyles to decide on the appropriate hairstyle for them. In this paper, an
approach is presented that uses K-means clustering to classify African women's
images. In order to identify potential facial clusters, Haarcascade is used for
feature-based training, and K-means clustering is applied for image
classification.
|
2306.06063 | Zhen Tan | Zhen Tan, Ruocheng Guo, Kaize Ding, Huan Liu | Virtual Node Tuning for Few-shot Node Classification | Accepted to KDD 2023 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot Node Classification (FSNC) is a challenge in graph representation
learning where only a few labeled nodes per class are available for training.
To tackle this issue, meta-learning has been proposed to transfer structural
knowledge from base classes with abundant labels to target novel classes.
However, existing solutions become ineffective or inapplicable when base
classes have no or limited labeled nodes. To address this challenge, we propose
an innovative method dubbed Virtual Node Tuning (VNT). Our approach utilizes a
pretrained graph transformer as the encoder and injects virtual nodes as soft
prompts in the embedding space, which can be optimized with few-shot labels in
novel classes to modulate node embeddings for each specific FSNC task. A unique
feature of VNT is that, by incorporating a Graph-based Pseudo Prompt Evolution
(GPPE) module, VNT-GPPE can handle scenarios with sparse labels in base
classes. Experimental results on four datasets demonstrate the superiority of
the proposed approach in addressing FSNC with unlabeled or sparsely labeled
base classes, outperforming existing state-of-the-art methods and even fully
supervised baselines.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 17:38:22 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Tan",
"Zhen",
""
],
[
"Guo",
"Ruocheng",
""
],
[
"Ding",
"Kaize",
""
],
[
"Liu",
"Huan",
""
]
] | TITLE: Virtual Node Tuning for Few-shot Node Classification
ABSTRACT: Few-shot Node Classification (FSNC) is a challenge in graph representation
learning where only a few labeled nodes per class are available for training.
To tackle this issue, meta-learning has been proposed to transfer structural
knowledge from base classes with abundant labels to target novel classes.
However, existing solutions become ineffective or inapplicable when base
classes have no or limited labeled nodes. To address this challenge, we propose
an innovative method dubbed Virtual Node Tuning (VNT). Our approach utilizes a
pretrained graph transformer as the encoder and injects virtual nodes as soft
prompts in the embedding space, which can be optimized with few-shot labels in
novel classes to modulate node embeddings for each specific FSNC task. A unique
feature of VNT is that, by incorporating a Graph-based Pseudo Prompt Evolution
(GPPE) module, VNT-GPPE can handle scenarios with sparse labels in base
classes. Experimental results on four datasets demonstrate the superiority of
the proposed approach in addressing FSNC with unlabeled or sparsely labeled
base classes, outperforming existing state-of-the-art methods and even fully
supervised baselines.
|
2306.06068 | Christian L\"owens | Christian L\"owens, Daniela Thyssens, Emma Andersson, Christina
Jenkins, Lars Schmidt-Thieme | DeepStay: Stay Region Extraction from Location Trajectories using Weak
Supervision | Paper under peer review | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Nowadays, mobile devices enable constant tracking of the user's position and
location trajectories can be used to infer personal points of interest (POIs)
like homes, workplaces, or stores. A common way to extract POIs is to first
identify spatio-temporal regions where a user spends a significant amount of
time, known as stay regions (SRs).
Common approaches to SR extraction are evaluated either solely unsupervised
or on a small-scale private dataset, as popular public datasets are unlabeled.
Most of these methods rely on hand-crafted features or thresholds and do not
learn beyond hyperparameter optimization. Therefore, we propose a weakly and
self-supervised transformer-based model called DeepStay, which is trained on
location trajectories to predict stay regions. To the best of our knowledge,
this is the first approach based on deep learning and the first approach that
is evaluated on a public, labeled dataset. Our SR extraction method outperforms
state-of-the-art methods. In addition, we conducted a limited experiment on the
task of transportation mode detection from GPS trajectories using the same
architecture and achieved significantly higher scores than the
state-of-the-art. Our code is available at
https://github.com/christianll9/deepstay.
| [
{
"version": "v1",
"created": "Mon, 5 Jun 2023 11:16:47 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Löwens",
"Christian",
""
],
[
"Thyssens",
"Daniela",
""
],
[
"Andersson",
"Emma",
""
],
[
"Jenkins",
"Christina",
""
],
[
"Schmidt-Thieme",
"Lars",
""
]
] | TITLE: DeepStay: Stay Region Extraction from Location Trajectories using Weak
Supervision
ABSTRACT: Nowadays, mobile devices enable constant tracking of the user's position and
location trajectories can be used to infer personal points of interest (POIs)
like homes, workplaces, or stores. A common way to extract POIs is to first
identify spatio-temporal regions where a user spends a significant amount of
time, known as stay regions (SRs).
Common approaches to SR extraction are evaluated either solely unsupervised
or on a small-scale private dataset, as popular public datasets are unlabeled.
Most of these methods rely on hand-crafted features or thresholds and do not
learn beyond hyperparameter optimization. Therefore, we propose a weakly and
self-supervised transformer-based model called DeepStay, which is trained on
location trajectories to predict stay regions. To the best of our knowledge,
this is the first approach based on deep learning and the first approach that
is evaluated on a public, labeled dataset. Our SR extraction method outperforms
state-of-the-art methods. In addition, we conducted a limited experiment on the
task of transportation mode detection from GPS trajectories using the same
architecture and achieved significantly higher scores than the
state-of-the-art. Our code is available at
https://github.com/christianll9/deepstay.
|
2306.06074 | Usman Nazir | Usman Nazir, Muhammad Ahmad Waseem, Falak Sher Khan, Rabia Saeed, Syed
Muhammad Hasan, Momin Uppal, Zubair Khalid | Improved flood mapping for efficient policy design by fusion of
Sentinel-1, Sentinel-2, and Landsat-9 imagery to identify population and
infrastructure exposed to floods | IEEE IGARSS 2023 | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | A reliable yet inexpensive tool for the estimation of flood water spread is
conducive for efficient disaster management. The application of optical and SAR
imagery in tandem provides a means of extended availability and enhanced
reliability of flood mapping. We propose a methodology to merge these two types
of imagery into a common data space and demonstrate its use in the
identification of affected populations and infrastructure for the 2022 floods
in Pakistan. The merging of optical and SAR data provides us with improved
observations in cloud-prone regions; that is then used to gain additional
insights into flood mapping applications. The use of open source datasets from
WorldPop and OSM for population and roads respectively makes the exercise
globally replicable. The integration of flood maps with spatial data on
population and infrastructure facilitates informed policy design. We have shown
that within the top five flood-affected districts in Sindh province, Pakistan,
the affected population accounts for 31 %, while the length of affected roads
measures 1410.25 km out of a total of 7537.96 km.
| [
{
"version": "v1",
"created": "Wed, 31 May 2023 20:46:06 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Nazir",
"Usman",
""
],
[
"Waseem",
"Muhammad Ahmad",
""
],
[
"Khan",
"Falak Sher",
""
],
[
"Saeed",
"Rabia",
""
],
[
"Hasan",
"Syed Muhammad",
""
],
[
"Uppal",
"Momin",
""
],
[
"Khalid",
"Zubair",
""
]
] | TITLE: Improved flood mapping for efficient policy design by fusion of
Sentinel-1, Sentinel-2, and Landsat-9 imagery to identify population and
infrastructure exposed to floods
ABSTRACT: A reliable yet inexpensive tool for the estimation of flood water spread is
conducive for efficient disaster management. The application of optical and SAR
imagery in tandem provides a means of extended availability and enhanced
reliability of flood mapping. We propose a methodology to merge these two types
of imagery into a common data space and demonstrate its use in the
identification of affected populations and infrastructure for the 2022 floods
in Pakistan. The merging of optical and SAR data provides us with improved
observations in cloud-prone regions; that is then used to gain additional
insights into flood mapping applications. The use of open source datasets from
WorldPop and OSM for population and roads respectively makes the exercise
globally replicable. The integration of flood maps with spatial data on
population and infrastructure facilitates informed policy design. We have shown
that within the top five flood-affected districts in Sindh province, Pakistan,
the affected population accounts for 31 %, while the length of affected roads
measures 1410.25 km out of a total of 7537.96 km.
|
2306.06078 | Haoxiang Yu | Haoxiang Yu, Jingyi An, Evan King, Edison Thomaz, Christine Julien | Cheating off your neighbors: Improving activity recognition through
corroboration | null | null | null | null | cs.CV cs.HC cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding the complexity of human activities solely through an
individual's data can be challenging. However, in many situations, surrounding
individuals are likely performing similar activities, while existing human
activity recognition approaches focus almost exclusively on individual
measurements and largely ignore the context of the activity. Consider two
activities: attending a small group meeting and working at an office desk. From
solely an individual's perspective, it can be difficult to differentiate
between these activities as they may appear very similar, even though they are
markedly different. Yet, by observing others nearby, it can be possible to
distinguish between these activities. In this paper, we propose an approach to
enhance the prediction accuracy of an individual's activities by incorporating
insights from surrounding individuals. We have collected a real-world dataset
from 20 participants with over 58 hours of data including activities such as
attending lectures, having meetings, working in the office, and eating
together. Compared to observing a single person in isolation, our proposed
approach significantly improves accuracy. We regard this work as a first step
in collaborative activity recognition, opening new possibilities for
understanding human activity in group settings.
| [
{
"version": "v1",
"created": "Sat, 27 May 2023 04:03:15 GMT"
}
] | 2023-06-12T00:00:00 | [
[
"Yu",
"Haoxiang",
""
],
[
"An",
"Jingyi",
""
],
[
"King",
"Evan",
""
],
[
"Thomaz",
"Edison",
""
],
[
"Julien",
"Christine",
""
]
] | TITLE: Cheating off your neighbors: Improving activity recognition through
corroboration
ABSTRACT: Understanding the complexity of human activities solely through an
individual's data can be challenging. However, in many situations, surrounding
individuals are likely performing similar activities, while existing human
activity recognition approaches focus almost exclusively on individual
measurements and largely ignore the context of the activity. Consider two
activities: attending a small group meeting and working at an office desk. From
solely an individual's perspective, it can be difficult to differentiate
between these activities as they may appear very similar, even though they are
markedly different. Yet, by observing others nearby, it can be possible to
distinguish between these activities. In this paper, we propose an approach to
enhance the prediction accuracy of an individual's activities by incorporating
insights from surrounding individuals. We have collected a real-world dataset
from 20 participants with over 58 hours of data including activities such as
attending lectures, having meetings, working in the office, and eating
together. Compared to observing a single person in isolation, our proposed
approach significantly improves accuracy. We regard this work as a first step
in collaborative activity recognition, opening new possibilities for
understanding human activity in group settings.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.