text
stringlengths 1
3.65k
| source
stringlengths 15
79
|
|---|---|
we consider data transmission through a time - selective, correlated ( first - order markov ) rayleigh fading channel subject to an average power constraint. the channel is estimated at the receiver with a pilot signal, and the estimate is fed back to the transmitter. the estimate is used for coherent demodulation, and to adapt the data and pilot powers. we explicitly determine the optimal pilot and data power control policies in a continuous - time limit where the channel state evolves as an ornstein - uhlenbeck diffusion process, and is estimated by a kalman filter at the receiver. the optimal pilot policy switches between zero and the maximum ( peak - constrained ) value ( ` ` bang - bang ' ' control ), and approximates the optimal discrete - time policy at low signal - to - noise ratios ( equivalently, large bandwidths ). the switching boundary is defined in terms of the system state ( estimated channel mean and associated error variance ), and can be explicitly computed. under the optimal policy, the transmitter conserves power by decreasing the training power when the channel is faded, thereby increasing the data rate. numerical results show a significant increase in achievable rate due to the adaptive training scheme with feedback, relative to constant ( non - adaptive ) training, which does not require feedback. the gain is more pronounced at relatively low snrs and with fast fading. results are further verified through monte carlo simulations.
|
arxiv:0908.1264
|
simultaneous simulation of hxr and mw emission with the same populations of electrons is still a great challenge for interpretation of observations in real events. in this paper we apply the fp kinetic model of precipitation of electron beam with energy range from 12 kev to 1. 2 mev to the interpretation of x - ray and microwave emissions observed in the flare of 10 march 2001. methods. the theoretical hxr and mw emissions were calculated by using the distribution functions of electron beams found by solving time - dependent fokker - planck approach in a converging magnetic field ( zharkova at al., 2010 ; kuznetsov and zharkova, 2010 ) for anisotropic scattering of beam electrons on the ambient particles in coloumb collisions and ohmic losses. the simultaneous observed hxr photon spectra and frequency distribution of mw emission and polarization were fit by those simulated from fp models which include the effects of electric field induced by beam electrons and precipitation into a converging magnetic loop. magnetic field strengths in the footpoints on the photosphere were updated with newly calibrated soho / mdi data. the observed hxr energy spectrum above 10 kev is shown to be a double power law which was fit precisely by the photon hxr spectrum simulated for the model including the self - induced electric field but without magnetic convergence. the mw emission simulated for different models of electron precipitation revealed a better fit to the observed distribution at higher frequencies for the models combining collisions and electric field effects with a moderate magnetic field convergence of 2. the mw simulations were able to reproduce closely the main features of the mw emission observed at higher frequencies.
|
arxiv:1105.3508
|
the in - medium properties of the vector mesons are known to be modified significantly in hot and dense hadronic matter due to vacuum polarisation effects from the baryon sector in the walecka model. the vector meson mass drops significantly in the medium due to the effects of the dirac sea. in the variational approach adopted in the present paper, these effects are taken into account through a realignment of the ground state with baryon condensates. such a realignment of the ground state becomes equivalent to summing of the baryonic tadpole diagrams in the relativistic hartree approximation ( rha ). the approximation scheme adopted here goes beyond rha to include quantum effects from the scalar meson and is nonperturbative and self - - consistent. it includes multiloop effects, thus corresponding to a different approximation as compared to the one loop approximation of including scalar field quantum corrections. in the present work, we study the properties of the vector mesons in the hot and dense matter as modified due to such quantum correction effects from the baryon as well as scalar meson sectors. these medium modifications of the properties of the vector mesons are reflected, through the shifting and broadening of the respective peaks, in the low mass dilepton spectra. there is broadening of the peaks due to corrections from scalar meson quantum effects as compared to the relativistic hartree approximation. it is seen to be rather prominent for the $ \ omega $ meson in the invariant mass plot.
|
arxiv:nucl-th/0201042
|
the map - reduce computing framework rose to prominence with datasets of such size that dozens of machines on a single cluster were needed for individual jobs. as datasets approach the exabyte scale, a single job may need distributed processing not only on multiple machines, but on multiple clusters. we consider a scheduling problem to minimize weighted average completion time of n jobs on m distributed clusters of parallel machines. in keeping with the scale of the problems motivating this work, we assume that ( 1 ) each job is divided into m " subjobs " and ( 2 ) distinct subjobs of a given job may be processed concurrently. when each cluster is a single machine, this is the np - hard concurrent open shop problem. a clear limitation of such a model is that a serial processing assumption sidesteps the issue of how different tasks of a given subjob might be processed in parallel. our algorithms explicitly model clusters as pools of resources and effectively overcome this issue. under a variety of parameter settings, we develop two constant factor approximation algorithms for this problem. the first algorithm uses an lp relaxation tailored to this problem from prior work. this lp - based algorithm provides strong performance guarantees. our second algorithm exploits a surprisingly simple mapping to the special case of one machine per cluster. this mapping - based algorithm is combinatorial and extremely fast. these are the first constant factor approximations for this problem.
|
arxiv:1610.09058
|
byzantine fault tolerant ( bft ) protocols play a pivotal role in blockchain technology. as the deployment of such systems extends to wide - area networks, the scalability of bft protocols becomes a critical concern. optimizations that assign specific roles to individual replicas can significantly improve the performance of bft systems. however, such role assignment is highly sensitive to faults, potentially undermining the optimizations effectiveness. to address these challenges, we present smartlog, a logging framework for collecting and analyzing metrics that help to assign roles in globally distributed systems, despite the presence of faults. smartlog presents local measurements in global data structures, to enable consistent decisions and hold replicas accountable if they do not perform according to their reported measurements. we apply smartlog to kauri, an optimization using randomly composed tree overlays. smartlog finds robust and low - latency tree configurations under adverse conditions.
|
arxiv:2502.15428
|
pre - trained language models ( plms ) like bert have made great progress in nlp. news articles usually contain rich textual information, and plms have the potentials to enhance news text modeling for various intelligent news applications like news recommendation and retrieval. however, most existing plms are in huge size with hundreds of millions of parameters. many online news applications need to serve millions of users with low latency tolerance, which poses huge challenges to incorporating plms in these scenarios. knowledge distillation techniques can compress a large plm into a much smaller one and meanwhile keeps good performance. however, existing language models are pre - trained and distilled on general corpus like wikipedia, which has some gaps with the news domain and may be suboptimal for news intelligence. in this paper, we propose newsbert, which can distill plms for efficient and effective news intelligence. in our approach, we design a teacher - student joint learning and distillation framework to collaboratively learn both teacher and student models, where the student model can learn from the learning experience of the teacher model. in addition, we propose a momentum distillation method by incorporating the gradients of teacher model into the update of student model to better transfer useful knowledge learned by the teacher model. extensive experiments on two real - world datasets with three tasks show that newsbert can effectively improve the model performance in various intelligent news applications with much smaller models.
|
arxiv:2102.04887
|
we compare a gromov hyperbolic metric with the hyperbolic metric in the unit ball or in the upper half space, and prove sharp comparison inequalities between the gromov hyperbolic metric and some hyperbolic type metrics. we also obtain several sharp distortion inequalities for the gromov hyperbolic metric under some families of m \ " { o } bius transformations.
|
arxiv:2006.04087
|
we consider physical processes caused by the twisted photons for a wide range of energy scales, from optical ( ev ) to nuclear ( mev ) to high - energy gamma - rays ( tev ). we demonstrate that in order to satisfy angular momentum conservation, absorption of a twisted photon leads to a transverse recoil of the final particle or a system of particles leading to increased threshold energy required for the reaction to proceed. modification of the threshold energy is predicted for ( a ) photo - absorption on colds trapped ions of $ ^ { 40 } $ ca, along with emerging new transverse - motion sidebands, ( b ) photo - disintegration of deuterium and ( c ) photo - production of electron - positron pairs in astrophysics environment.
|
arxiv:2007.05816
|
animatable 3d reconstruction has significant applications across various fields, primarily relying on artists ' handcraft creation. recently, some studies have successfully constructed animatable 3d models from monocular videos. however, these approaches require sufficient view coverage of the object within the input video and typically necessitate significant time and computational costs for training and rendering. this limitation restricts the practical applications. in this work, we propose a method to build animatable 3d gaussian splatting from monocular video with diffusion priors. the 3d gaussian representations significantly accelerate the training and rendering process, and the diffusion priors allow the method to learn 3d models with limited viewpoints. we also present the rigid regularization to enhance the utilization of the priors. we perform an extensive evaluation across various real - world videos, demonstrating its superior performance compared to the current state - of - the - art methods.
|
arxiv:2403.11427
|
the noether charge method for defining the hamiltonian of a diffeomorphism - invariant field theory is applied to " einstein - aether " theory, in which gravity couples to a dynamical, timelike, unit - norm vector field. using the method, expressions are obtained for the total energy, momentum, and angular momentum of an einstein - aether space - time. the method is also used to discuss the mechanics of einstein - aether black holes. the derivation of wald, and iyer and wald, of the first law of black hole thermodynamics fails for this theory, because the unit vector is necessarily singular at the bifurcation surface of the killing horizon. a general identity relating variations of energy and angular momentum to a surface integral at the horizon is obtained, but a thermodynamic interpretation, including a definitive expression for the black hole entropy, is not found.
|
arxiv:gr-qc/0509121
|
graph motif, defined as the microstructure that appears repeatedly in a large graph, reveals important topological characteristics of the large graph and has gained increasing attention in power system analysis regarding reliability, vulnerability and resiliency. however, searching motifs within the large - scale power system is extremely computationally challenging and even infeasible, which undermines the value of motif analysis in practice. in this paper, we introduce a two - stage ai - powered motif mining method to enable efficient and wide - range motif analysis in power systems. in the first stage, a representation learning method with specially designed network structure and loss function is proposed to achieve ordered embedding for the power system topology, simplifying the subgraph isomorphic problem into a vector comparison problem. in the second stage, under the guidance of the ordered embedding space, a greedy - search - based motif growing algorithm is introduced to quickly obtain the motifs without traversal searching. a case study based on a power system database including 61 circuit models demonstrates the effectiveness of the proposed method.
|
arxiv:2412.05957
|
referring video object segmentation aims to predict foreground labels for objects referred by natural language expressions in videos. previous methods either depend on 3d convnets or incorporate additional 2d convnets as encoders to extract mixed spatial - temporal features. however, these methods suffer from spatial misalignment or false distractors due to delayed and implicit spatial - temporal interaction occurring in the decoding phase. to tackle these limitations, we propose a language - bridged duplex transfer ( lbdt ) module which utilizes language as an intermediary bridge to accomplish explicit and adaptive spatial - temporal interaction earlier in the encoding phase. concretely, cross - modal attention is performed among the temporal encoder, referring words and the spatial encoder to aggregate and transfer language - relevant motion and appearance information. in addition, we also propose a bilateral channel activation ( bca ) module in the decoding phase for further denoising and highlighting the spatial - temporal consistent features via channel - wise activation. extensive experiments show our method achieves new state - of - the - art performances on four popular benchmarks with 6. 8 % and 6. 9 % absolute ap gains on a2d sentences and j - hmdb sentences respectively, while consuming around 7x less computational overhead.
|
arxiv:2206.03789
|
a separate pool of p { \ displaystyle p } balls is : p − 1 p ( k b ) ( n − k k − b ) ( n k ) { \ displaystyle { p - 1 \ over p } { k \ choose b } { n - k \ choose k - b } \ over { n \ choose k } } = = minimum number of tickets for a match = = it is a hard ( and often open ) problem to calculate the minimum number of tickets one needs to purchase to guarantee that at least one of these tickets matches at least 2 numbers. in the 5 - from - 90 lotto, the minimum number of tickets that can guarantee a ticket with at least 2 matches is 100. = = coincidences involving lottery numbers = = coincidences in lottery drawings often capture our imagination and can make news headlines as they seemingly highlight patterns in what should be entirely random outcomes. for example, repeated numbers appearing across different draws may appear on the surface to be too implausible to be by pure chance. for instance, on september 6, 2009, the six numbers 4, 15, 23, 24, 35, and 42 were drawn from 49 in the bulgarian national 6 / 49 lottery, and in the very next drawing on september 10th, the same six numbers were drawn again. lottery mathematics can be used to analyze these extraordinary events. = = information theoretic results = = as a discrete probability space, the probability of any particular lottery outcome is atomic, meaning it is greater than zero. therefore, the probability of any event is the sum of probabilities of the outcomes of the event. this makes it easy to calculate quantities of interest from information theory. for example, the information content of any event is easy to calculate, by the formula i ( e ) : = − log [ pr ( e ) ] = − log ( p ). { \ displaystyle \ operatorname { i } ( e ) : = - \ log { \ left [ \ pr { \ left ( e \ right ) } \ right ] } = - \ log { \ left ( p \ right ) }. } in particular, the information content of outcome x { \ displaystyle x } of discrete random variable x { \ displaystyle x } is i x ( x ) : = − log [ p x ( x ) ] = log ( 1 p x ( x ) ). { \ displaystyle \ operatorname { i } _ { x } ( x ) : = - \ log { \ left [ p _
|
https://en.wikipedia.org/wiki/Lottery_mathematics
|
denoising diffusion probabilistic models ( ddpms ) have gained great attention in adversarial purification. current diffusion - based works focus on designing effective condition - guided mechanisms while ignoring a fundamental problem, i. e., the original ddpm sampling is intended for stable generation, which may not be the optimal solution for adversarial purification. inspired by the stability of the denoising diffusion implicit model ( ddim ), we propose an opposite sampling scheme called random sampling. in brief, random sampling will sample from a random noisy space during each diffusion process, while ddpm and ddim sampling will continuously sample from the adjacent or original noisy space. thus, random sampling obtains more randomness and achieves stronger robustness against adversarial attacks. correspondingly, we also introduce a novel mediator conditional guidance to guarantee the consistency of the prediction under the purified image and clean image input. to expand awareness of guided diffusion purification, we conduct a detailed evaluation with different sampling methods and our random sampling achieves an impressive improvement in multiple settings. leveraging mediator - guided random sampling, we also establish a baseline method named diffap, which significantly outperforms state - of - the - art ( sota ) approaches in performance and defensive stability. remarkably, under strong attack, our diffap even achieves a more than 20 % robustness advantage with 10 $ \ times $ sampling acceleration.
|
arxiv:2411.18956
|
in this paper we construct for every integer $ n > 1 $ a complex manifold of dimension $ n $ which is exhausted by an increasing sequence of biholomorphic images of $ \ mathbb c ^ n $ ( i. e., a long $ \ mathbb c ^ n $ ), but it does not admit any nonconstant holomorphic or plurisubharmonic functions. furthermore, we introduce new biholomorphic invariants of a complex manifold $ x $, the stable core and the strongly stable core, that are based on the long term behavior of hulls of compact sets with respect to an exhaustion of $ x $. we show that every compact polynomially convex set $ b \ subset \ mathbb c ^ n $ which is the closure of its interior is the strongly stable core of a long $ \ mathbb c ^ n $ ; in particular, biholomorphically nonequivalent sets give rise to nonequivalent long $ \ mathbb c ^ n $ ' s. furthermore, for any open set $ u \ subset \ mathbb c ^ n $ there exists a long $ \ mathbb c ^ n $ whose stable core is dense in $ u $. it follows that for any $ n > 1 $ there is a continuum of pairwise nonequivalent long $ \ mathbb c ^ n $ ' s with no nonconstant plurisubharmonic functions and no nontrivial holomorphic automorphisms. these results answer several long standing open problems.
|
arxiv:1511.05075
|
a study of 12 ultracompact hii regions was conducted to probe the physical conditions and kinematics in the inner envelopes of the molecular clumps harboring them. the apex telescope was used to observe the sources in the co ( 4 - 3 ) and 13co ( 8 - 7 ) lines. line intensities were modeled with the ratran radiative transfer code using power laws for the density and temperature to describe the physical structure of the clumps. all sources were detected in both lines. the optically thick co ( 4 - 3 ) line shows predominantly blue skewed profiles reminiscent of infall. line intensities can be reproduced well using the physical structure of the clumps taken from the literature. the optically thick line profiles show that co is a sensitive tracer of ongoing infall in the outer envelopes of clumps harboring ultracompact hii regions and hot molecular cores.
|
arxiv:astro-ph/0605668
|
internet of things ( iot ) systems are recently being employed in various types of critical infrastructure, including integrated rescue systems, healthcare, defence, energy and other fields. recently, the security and safety of iot systems, in general, has been questioned by a number of studies. raised concerns do not relate to the iot technology in principle but to poor engineering practices that are mostly preventable. in critical infrastructure, demand for safety and security is strongly present and justifies a discussion about the general resilience of iot systems. in this context, resilience includes system resistance to cyberattacks and its stability to operating conditions and system reliability and safety in terms of present flaws. in this paper, we discuss relevant factors impacting the resilience of iot systems in the critical infrastructure and suggest possible countermeasures and actions mitigate the potential effects of these factors. contrary to the previous work, an unique critical system model - based testing viewpoint is taken in this analysis.
|
arxiv:2205.13576
|
we obtain a generalization of the low theorem for non - abelian boson emission in collision of scalar and gauge vector particles and its extension to high energy collisions for small transverse momenta of produced particles. we demonstrate that in the case of particles with spin the direct extension the low formula to high energy is in contradiction with the correct amplitude factorization behavior. consideration of different kinematical regions and use of methods of dual models allows us to separate contributions of intermediate excited states and standard spin corrections in the low formulae. we show that the amplitude factorization occurs at high energy due to the contribution of the intermediate states which is additional to the gluon production amplitude for the scalar particle collision.
|
arxiv:hep-ph/9503428
|
we present a comprehensive mathematical study of the magneto - telluric ( mt ) method, on bounded domain in $ \ mathbb { r } ^ 3 $. we show that electrical conductivity and magnetic permeability, assumed to be $ c ^ 2 $, can be uniquely recovered from mt data measured on the boundary of the domain. the proof is based on the construction of complex geometric optics solutions. furthermore, we obtain a unique determination result in the case when the mt data are measured only on an open subset of the boundary. here, we assume that the part of the boundary inaccessible for measurements is a subset of a sphere.
|
arxiv:2012.00149
|
we provide sample complexity upper bounds for agnostically learning multivariate gaussians under the constraint of approximate differential privacy. these are the first finite sample upper bounds for general gaussians which do not impose restrictions on the parameters of the distribution. our bounds are near - optimal in the case when the covariance is known to be the identity, and conjectured to be near - optimal in the general case. from a technical standpoint, we provide analytic tools for arguing the existence of global " locally small " covers from local covers of the space. these are exploited using modifications of recent techniques for differentially private hypothesis selection. our techniques may prove useful for privately learning other distribution classes which do not possess a finite cover.
|
arxiv:2010.09929
|
significant scientific and technological progress in the field of spintronics is based on trilayer magnetic tunnel junction devices which principally rely on the physics of single barrier tunneling. while technologically relevant devices have been prototyped, the physics of single barrier tunneling poses ultimate limitations on the performance of magnetic tunnel junction devices. here, we propose a fresh route toward high performance magnetic tunnel junctions by making electronic analogs of optical phenomena such as anti - reflections and fabry - p \ ` erot resonances. the devices we propose feature anti - reflection enabled superlattice heterostructures sandwiched between the fixed and the free ferromagnets of the magnetic tunnel junction structure. our predictions are based on the non - equilibrium green ' s function spin transport formalism coupled self - consistently with the landau - lifshitz - gilbert - slonczewski equation. owing to the physics of bandpass spin filtering in the bandpass superlattice magnetic tunnel junction device, we demonstrate an ultra - high boost in the tunnel magneto - resistance ( tmr $ \ approx5 \ times10 ^ 4 \ % $ ) and nearly 92 % suppression of spin transfer torque switching bias in comparison to a traditional trilayer magnetic tunnel junction device. we rationalize improvised spin transfer torque switching via analysis of the slonczewski spin current transmission spectra. the proof of concepts presented here can lead to next - generation spintronics device design harvesting the rich physics of superlattice heterostructures and exploiting spintronic analogs of optical phenomena.
|
arxiv:1801.09409
|
the ratio set of a set of positive integers $ a $ is defined as $ r ( a ) : = \ { a / b : a, b \ in a \ } $. the study of the denseness of $ r ( a ) $ in the set of positive real numbers is a classical topic and, more recently, the denseness in the set of $ p $ - adic numbers $ \ mathbb { q } _ p $ has also been investigated. let $ a _ 1, \ ldots, a _ k $ be a partition of $ \ mathbb { n } $ into $ k $ sets. we prove that for all prime numbers $ p $ but at most $ \ lfloor \ log _ 2 k \ rfloor $ exceptions at least one of $ r ( a _ 1 ), \ ldots, r ( a _ k ) $ is dense in $ \ mathbb { q } _ p $. moreover, we show that for all prime numbers $ p $ but at most $ k - 1 $ exceptions at least one of $ a _ 1, \ ldots, a _ k $ is dense in $ \ mathbb { z } _ p $. both these results are optimal in the sense that there exist partitions $ a _ 1, \ ldots, a _ k $ having exactly $ \ lfloor \ log _ 2 k \ rfloor $, respectively $ k - 1 $, exceptional prime numbers ; and we give explicit constructions for them. furthermore, as a corollary, we answer negatively a question raised by garcia et al.
|
arxiv:1808.00374
|
recently, unsupervised local learning, based on hebb ' s idea that change in synaptic efficacy depends on the activity of the pre - and postsynaptic neuron only, has shown potential as an alternative training mechanism to backpropagation. unfortunately, hebbian learning remains experimental and rarely makes it way into standard deep learning frameworks. in this work, we investigate the potential of hebbian learning in the context of standard deep learning workflows. to this end, a framework for thorough and systematic evaluation of local learning rules in existing deep learning pipelines is proposed. using this framework, the potential of hebbian learned feature extractors for image classification is illustrated. in particular, the framework is used to expand the krotov - hopfield learning rule to standard convolutional neural networks without sacrificing accuracy compared to end - to - end backpropagation. the source code is available at https : / / github. com / joxis / pytorch - hebbian.
|
arxiv:2102.00428
|
all - dielectric resonant micro - and nano - structures made of the high - index dielectrics recently emerge as a promising sers platform which can complement or potentially replace the metal - based counterparts in routine sensing measurements. these unique structures combine the highly - tunable optical response and high field enhancement with the non - invasiveness, i. e., chemically non - perturbing the analyte, simple chemical modification and recyclability. meanwhile, the commercially competitive fabrication technologies for mass production of such structures are still missing. here, we attest a chemically inert black silicon ( b - si ) substrate consisting of randomly - arranged spiky mie resonators for a true non - invasive sers identification of the molecular fingerprints at low concentrations. based on comparative in - situ sers tracking of the para - aminothiophenol - to - 4, 4 ` dimercaptoazobenzene catalytic conversion on the bare and metal - coated b - si, we justify applicability of the metal - free b - si for the ultra - sensitive non - invasive sers detection at concentration level as low as 10 ^ - 6 m. we perform finite - difference time - domain calculations to reveal the electromagnetic enhancement provided by an isolated spiky si resonator in the visible spectral range. additional comparative sers studies of the patp - to - dmab conversion performed with a chemically active bare black copper oxide as well as sers detection of the slow daylight - driven patp - to - damp catalytic conversion in the aqueous methanol solution loaded with colloidal silver nanoparticles confirm the non - invasive sers performance of the all - dielectric crystalline b - si substrate. proposed sers substrate can be fabricated using simple scalable technology of plasma etching amenable on large substrate areas making such inexpensive all - dielectric substrates promising for routine sers applications, where the non - invasiveness is of mandatory importance.
|
arxiv:1802.09783
|
representation engineering ( repe ) is a novel paradigm for controlling the behavior of llms. unlike traditional approaches that modify inputs or fine - tune the model, repe directly manipulates the model ' s internal representations. as a result, it may offer more effective, interpretable, data - efficient, and flexible control over models ' behavior. we present the first comprehensive survey of repe for llms, reviewing the rapidly growing literature to address key questions : what repe methods exist and how do they differ? for what concepts and problems has repe been applied? what are the strengths and weaknesses of repe compared to other methods? to answer these, we propose a unified framework describing repe as a pipeline comprising representation identification, operationalization, and control. we posit that while repe methods offer significant potential, challenges remain, including managing multiple concepts, ensuring reliability, and preserving models ' performance. towards improving repe, we identify opportunities for experimental and methodological improvements and construct a guide for best practices.
|
arxiv:2502.19649
|
wire - grid polarisers are versatile and scalable components which can be engineered to achieve small sizes and extremely high extinction ratios. yet the measured performances are always significantly below the predicted values obtained from numerical simulations. here we report on a detailed comparison between theoretical and experimental performances. we show that the discrepancy can be explained by the true shape of the plasmonic structures. taking into account the fabrication details, a new optimisation model enables us to achieve excellent agreement with the observed response and to re - optimise the grating parameters to ensure experimental extinction ratios well above 1, 000 at 850 nm.
|
arxiv:1508.04355
|
by using the valon model, we calculate the transverse momentum distribution functions, tmds, inside the nucleon. tmds indicate the probability to find partons with spin aligned ( anti - aligned ) to the transversely polarized nucleon. the results are in good agreements with all available experimental data and also global fits.
|
arxiv:1401.1295
|
the effective low - energy models of the hubbard model are usually derived from perturbation theory. here we derive the effective model of the hubbard model in spin space and t - j space using a unitary transformation from numerical optimization. we represent the hamiltonian as matrix product state ( mpo ) and represent the unitary transformation using gates according to tensor network methods. we obtain this unitary transformation by optimizing the unitary transformation between the ground state of the hubbard model and the projection of the hubbard model ground state into spin space and t - j space. the unitary transformation we get from numerical optimization yields effective models that are in line with perturbation theories. this numerical optimization method starting from ground state provides another approach to analyze effective low - energy models of strongly correlated electron systems.
|
arxiv:1908.03979
|
end - to - end speech translation ( e2e - st ) has received increasing attention due to the potential of its less error propagation, lower latency, and fewer parameters. however, the effectiveness of neural - based approaches to this task is severely limited by the available training corpus, especially for domain adaptation where in - domain triplet training data is scarce or nonexistent. in this paper, we propose a novel non - parametric method that leverages domain - specific text translation corpus to achieve domain adaptation for the e2e - st system. to this end, we first incorporate an additional encoder into the pre - trained e2e - st model to realize text translation modelling, and then unify the decoder ' s output representation for text and speech translation tasks by reducing the correspondent representation mismatch in available triplet training data. during domain adaptation, a k - nearest - neighbor ( knn ) classifier is introduced to produce the final translation distribution using the external datastore built by the domain - specific text translation corpus, while the universal output representation is adopted to perform a similarity search. experiments on the europarl - st benchmark demonstrate that when in - domain text translation data is involved only, our proposed approach significantly improves baseline by 12. 82 bleu on average in all translation directions, even outperforming the strong in - domain fine - tuning method.
|
arxiv:2205.11211
|
an effect of overlapping of exhausted regions around droplets of a new phase is analyzed. several approximation to take overlapping into account are suggested. the small parameter responsible for convergence is extracted. the multi particle effects of overlapping are described.
|
arxiv:0709.1724
|
the outer disk of the lmc is studied in order to unveil clues about its formation and evolution. complementing our previous studies in innermost fields ( 3 < r < 7 kpc ), we obtained deep color magnitude diagrams in 6 fields with radius from 5. 2 to 9. 2 kpc. the comparison with isochrones shows that while the oldest population is approximately coeval in all fields, the age of the youngest populations increases with increasing radius. low - resolution spectroscopy in the infrared caii triplet region has been obtained for about 150 stars near the tip red giant branch in the same fields. radial velocities and stellar metallicities have been obtained from these spectra. the metallicity distribution of each field has been analyzed together with those previously studied. the metal content of the most metal - poor objects, which are also the oldest according to the derived age - metallicity relationships, is similar in all fields independently of the radius. however, while the metallicity of the most metal - rich objects measured, which are the youngest ones, remains constant in the inner 6 kpc, it decreases with increasing radius from there off. the same is true for the mean metallicity. according to the derived age - metallicity relationships, which are consistent with being the same in all fields, this result may be interpreted as an outside - in formation scheme in opposition with the inside - out scenario predicted by lcdm cosmology for a galaxy like the lmc. the analysis of the radial velocities of our sample of giants shows that they follow a rotational cold disk kinematics. the velocity dispersion increases as metallicity decreases indicating that the most metal - poor / oldest objects are distributed in a thicker disk than the most metal - rich / youngest ones in agreement with the findings in other disks such as that of the milky way. they do not seem to be part of a hot halo, if one exists in the lmc.
|
arxiv:1106.3418
|
for a family $ \ mathcal { f } $ of subsets of a finite set, define $ \ mathcal { d } ( \ mathcal { f } ) = \ { f \ setminus f ' : f, f ' \ in \ mathcal { f } \ } $. a family $ \ mathcal { f } $ is called intersecting if $ f \ cap f ' \ not = \ emptyset $ for all $ f, f ' \ in \ mathcal { f } $. frankl \ cite { frankl } showed that for a $ k $ - uniform intersecting family $ \ mathcal { f } \ subset { [ n ] \ choose k } $ with $ n \ ge k ( k + 3 ) $, $ | \ mathcal { d } ( \ mathcal { f } ) | $ reaches the maximum if and only if $ \ mathcal { f } $ is a $ k $ - uniform full star. later, frankl - kiselev - kupavskii \ cite { fkk } improved the bound $ n \ ge k ( k + 3 ) $ in the above result of frankl \ cite { frankl } to $ n \ ge 50klnk $ for $ k \ ge 50 $. for $ 2k < n < 4k $, frankl - kiselev - kupavskii \ cite { fkk } showed that there exists a $ k $ - uniform family $ \ mathcal { f } \ subset { [ n ] \ choose k } $ such that $ | \ mathcal { d } ( \ mathcal { f } ) | $ is larger than $ | \ mathcal { d } ( \ mathcal { s } ) | $, where $ \ mathcal { s } $ is a full star. this result left the case $ n = 2k $ open and we show that $ \ mathcal { d } ( \ mathcal { f } ) $ can be ` full ' for some $ \ mathcal { f } \ subset { [ n ] \ choose k } $. it is clear that for an intersecting family $ \ mathcal { f } \ subset { [ n ] \ choose k } $, $ \ mathcal { d } ( \ mathcal { f } ) \ subseteq \ cup _ { j = 0 } ^ { k - 1 } { [ n ] \ choose j } $. we say that a $ k
|
arxiv:2410.23723
|
the results of a single - crystal x - ray - diffraction study of the evolution of crystal structures of vi3 with temperature with emphasis on phase transitions are presented. some related specific - heat and magnetization data are included. the existence of the room - temperature trigonal crystal structure r - 3 ( 148 ) has been confirmed. upon cooling, vi3 undergoes a structural phase transition to a monoclinic phase at ts ~ 79 k. ts is reduced in magnetic fields applied along the trigonal c - axis. when vi3 becomes ferromagnetic at tfm1 ~ 50 k, magnetostriction - induced changes of the monoclinic - structure parameters are observed. upon further cooling, the monoclinic structure transforms into a triclinic variant at 32 k which is most likely occurring in conjunction with the previously reported transformation of the ferromagnetic structure. the observed phenomena are preliminarily attributed to strong magnetoelastic interactions.
|
arxiv:1909.04459
|
in this paper, we examine a scenario in which late - time cosmic acceleration might arise due to the coupling between baryonic matter and dark matter without the presence of extra degrees of freedom. in this case, one can obtain late - time acceleration in jordan frame and not in einstein frame. we consider two different forms of parametrization of the coupling function, and put constraints on the model parameters by using an integrated datasets of hubble parameter, type ia supernova and baryon acoustic oscillations. the models under consideration are consistent with the observations. in addition, we perform the statefinder and $ om $ diagnostics, and show that the models exhibit a distinctive behavior due to the phantom characteristic in future which is a generic feature of the underlying scenario.
|
arxiv:1711.06321
|
condition imposed on the nonlinear terms of a nonlinear diffusion equation with { r } obin boundary condition is the main focus of this paper. the degenerate parabolic equations, such as the { s } tefan problem, the { h } ele - - { s } haw problem, the porous medium equation and the fast diffusion equation, are included in this class. by characterizing this class of equations as an asymptotic limit of the { c } ahn - - { h } illiard systems, the growth condition of the nonlinear term can be improved. in this paper, the existence and uniqueness of the solution are proved. from the physical view point, it is natural that, the { c } ahn - - { h } illiard system is treated under the homogeneous { n } eumann boundary condition. therefore, the { c } ahn - - { h } illiard system subject to the { r } obin boundary condition looks like pointless. however, at some level of approximation, it makes sense to characterize the nonlinear diffusion equations.
|
arxiv:1802.02755
|
for a connected smooth projective curve $ x $ of genus $ g $, global sections of any line bundle $ l $ with $ \ deg ( l ) \ geq 2g + 1 $ give an embedding of the curve into projective space. we consider an analogous statement for a berkovich skeleton in nonarchimedean geometry, in which projective space is replaced by tropical projective space, and an embedding is replaced by a homeomorphism onto its image preserving integral structures ( called a faithful tropicalization ). let $ k $ be an algebraically closed field which is complete with respect to a non - trivial nonarchimedean value. suppose that $ x $ is defined over $ k $ and has genus $ g \ geq 2 $ and that $ \ gamma $ is a skeleton ( that is allowed to have ends ) of the analytification $ x ^ { \ mathrm { an } } $ of $ x $ in the sense of berkovich. we show that if $ \ deg ( l ) \ geq 3g - 1 $, then global sections of $ l $ give a faithful tropicalization of $ \ gamma $ into tropical projective space. as an application, when $ y $ is a suitable affine curve, we describe the analytification $ y ^ { \ mathrm { an } } $ as the limit of tropicalizations of an effectively bounded degree.
|
arxiv:1612.01098
|
in the mssm with complex parameters, loop corrections to the decay of a stop into a bottom quark and a chargino can lead to a cp violating decay rate asymmetry. we calculate this asymmetry at full one - loop level and perform a detailed numerical study, analyzing the dependence on the parameters and complex phases involved. if the stop can decay into a gluino, the self - energy and the vertex correction dominate due to the strong coupling. it is shown that the vertex contribution is always suppressed. we therefore give a simple approximate formula for the asymmetry. we account for the constraints on the parameters coming from several experimental limits. asymmetries up to 25 percent are obtained. we also comment on the feasibility of measuring this asymmetry at the lhc.
|
arxiv:0912.4675
|
heterogeneous nanoscale particles released by cells known as extracellular vesicles ( evs ) are actively investigated for early disease detection1, monitoring2, and advanced therapeutics3. due to their extremely small size, the stable trapping of nano - sized evs using diffraction - limited optical tweezers4 has been met with challenges. plasmon - enhanced optical trapping can confine light to the nanoscale to generate tight trapping potentials. unfortunately, a long - standing challenge is that plasmonic tweezers have limited throughput and cannot provide rapid delivery and trapping of particles at plasmonic hotspots while precluding the intrinsic plasmon - induced photothermal heating effect at the same time. we report our original geometry - induced electrohydrodynamic tweezers ( get ) that generate multiple electrohydrodynamic potentials for the parallelized transport and trapping of single evs in parallel within seconds while enhancing the imaging of single trapped evs. we show that the integration of nanoscale plasmonic cavities at the center of each get trap results in the parallel placement of single evs near plasmonic cavities enabling instantaneous plasmon - enhanced optical trapping upon laser illumination without any detrimental heating effect for the first time. these non - invasive scalable hybrid nanotweezers open new horizons for high - throughput tether - free plasmon - enhanced single ev trapping and spectroscopy. other potential areas of impact include nanoplastics characterization, and scalable hybrid integration for quantum photonics.
|
arxiv:2302.07370
|
state - of - the - art physics experiments require high - resolution, low - noise, and low - threshold detectors to achieve competitive scientific results. however, experimental environments invariably introduce sources of noise, such as electrical interference or microphonics. the sources of this environmental noise can often be monitored by adding specially designed " auxiliary devices " ( e. g. microphones, accelerometers, seismometers, magnetometers, and antennae ). a model can then be constructed to predict the detector noise based on the auxiliary device information, which can then be subtracted from the true detector signal. here, we present a multivariate noise cancellation algorithm which can be used in a variety of settings to improve the performance of detectors using multiple auxiliary devices. to validate this approach, we apply it to simulated data to remove noise due to electromagnetic interference and microphonic vibrations. we then employ the algorithm to a cryogenic light detector in the laboratory and show an improvement in the detector performance. finally, we motivate the use of nonlinear terms to better model vibrational contributions to the noise in thermal detectors. we show a further improvement in the performance of a particular channel of the cuore detector when using the nonlinear algorithm in combination with optimal filtering techniques.
|
arxiv:2311.01131
|
in the last decade cherenkov telescopes on the ground and space - based \ gamma - ray instruments have identified a new class of high mass x - ray binaries ( hmxb ), whose emission is dominated by \ gamma rays. to date only five of these systems are known. all of them are detected by cherenkov telescopes in the tev energy range, while at gev energies there is still one ( hess j0632 + 057 ) that has no reported detection with the fermi - lat. a deep search for \ gamma - ray emission of hess j0632 + 057 has been performed using more than 3. 5 years of fermi - lat data. we discuss the results of this search and compare it to other \ gamma - ray binary systems.
|
arxiv:1211.6142
|
the fundamental question is how the hyperon plays a role in the nuclear structure. it is of particular importance, especially in the light mass region, to verify the structure change when $ \ lambda $ particle ( s ) is added to normal nuclei. the ground state of $ ^ { 8 } $ be has been know to have a well - developed $ \ alpha $ - - $ \ alpha $ cluster structure, whereasn $ ^ { 12 } $ c has a mixed structure of three $ \ alpha $ clusters and $ jj $ - coupling shell model, where $ \ alpha $ clusters are partially broken. adding $ \ lambda $ particle ( s ) could induce the structure change. we compare the be and c cases. using the antisymmetrized quasi - cluster model ( aqcm ), the $ \ alpha $ - cluster states and $ jj $ - coupling shell - model states of $ ^ 8 $ be and $ ^ { 12 } $ c are prepared on the same footing, and we add $ \ lambda $ particles. the cluster - shell competition in the ground state can be well described with this model. using aqcm, we calculate $ ^ 8 $ be, $ ^ { 9 } _ { \ lambda } $ be, $ ^ { 10 } _ { \ lambda \ lambda } $ be, $ ^ { 12 } $ c, $ ^ { 13 } _ { \ lambda } $ c, and $ ^ { 14 } _ { \ lambda \ lambda } $ c. by adding one or two $ \ lambda $ particle ( s ), the ground state of $ ^ { 12 } $ c approaches the $ jj $ - coupling shell model side. on the other hand, in the be case, although the $ \ lambda $ particle ( s ) shrinks the $ \ alpha $ - - $ \ alpha $ distance, the breaking effect of the cluster structure is rather limited. the spin - orbit interaction is the driving force of breaking the $ \ alpha $ clusters, and whether the glue - like effect of $ \ lambda $ particle ( s ) attracts the cluster inside the range of this interaction is crucial. in $ ^ { 14 } _ { \ lambda \ lambda } $ c, the breaking of $ \ alpha $ clusters in $ ^ { 12 } $ c is much enhanced by the addition of the $ \ lambda $ particles than the case of free $ ^ { 12 } $ c. we also found that breaking $ \ alpha $ clusters in the ground state of $ ^ { 14 } _ {
|
arxiv:2301.01878
|
vector boson scattering ( vbs ) and vector boson fusion ( vbf ) studies in pp collisions at 7 and 8 tev center of mass energy based on data recorded by the cms detector at the lhc in 2011 and 2012 are reported.
|
arxiv:1409.2990
|
training deep neural networks for image recognition often requires large - scale human annotated data. to reduce the reliance of deep neural solutions on labeled data, state - of - the - art semi - supervised methods have been proposed in the literature. nonetheless, the use of such semi - supervised methods has been quite rare in the field of facial expression recognition ( fer ). in this paper, we present a comprehensive study on recently proposed state - of - the - art semi - supervised learning methods in the context of fer. we conduct comparative study on eight semi - supervised learning methods, namely pi - model, pseudo - label, mean - teacher, vat, mixmatch, remixmatch, uda, and fixmatch, on three fer datasets ( fer13, raf - db, and affectnet ), when various amounts of labeled samples are used. we also compare the performance of these methods against fully - supervised training. our study shows that when training existing semi - supervised methods on as little as 250 labeled samples per class can yield comparable performances to that of fully - supervised methods trained on the full labeled datasets. to facilitate further research in this area, we make our code publicly available at : https : / / github. com / shuvenduroy / ssl _ fer
|
arxiv:2208.00544
|
viral kinetics have been extensively studied in the past through the use of spatially homogeneous ordinary differential equations describing the time evolution of the diseased state. however, spatial characteristics such as localized populations of dead cells might adversely affect the spread of infection, similar to the manner in which a counter - fire can stop a forest fire from spreading. in order to investigate the influence of spatial heterogeneities on viral spread, a simple 2 - d cellular automaton ( ca ) model of a viral infection has been developed. in this initial phase of the investigation, the ca model is validated against clinical immunological data for uncomplicated influenza a infections. our results will be shown and discussed.
|
arxiv:q-bio/0402012
|
we present photometry and spectroscopy of four hydrogen - poor luminous supernovae discovered during the two - month science commissioning and early operations of the zwicky transient facility ( ztf ) survey. three of these objects, sn2018bym ( ztf18aapgrxo ), sn2018avk ( ztf18aaisyyp ) and sn2018bgv ( ztf18aavrmcg ) resemble typical slsn - i spectroscopically, while sn2018don ( ztf18aajqcue ) may be an object similar to sn2007bi experiencing considerable host galaxy reddening, or an intrinsically long - lived, luminous and red sn ic. we analyze the light curves, spectra, and host galaxy properties of these four objects and put them in context of the population of slsn - i. sn2018bgv stands out as the fastest - rising slsn - i observed to date, with a rest - frame g - band rise time of just 10 days from explosion to peak - - if it is powered by magnetar spin - down, the implied ejecta mass is only ~ 1 m $ _ { \ odot } $. sn2018don also displays unusual properties - - in addition to its red colors and comparatively massive host galaxy, the light curve undergoes some of the strongest light curve undulations post - peak seen in a slsn - i, which we speculate may be due to interaction with circumstellar material. we discuss the promises and challenges of finding slsne in large - scale surveys like ztf given the observed diversity in the population.
|
arxiv:1910.02968
|
reflection at an interface separating two different media is a rather universal phenomenon which arises because of wave mismatching at the interface. by means of supersymmetric quantum mechanics methods, it is shown that a fully transparent interface can be realized, connecting two isospectral but different one - dimensional crystals. an example of reflectionless interface is presented for the sinusoidal ( mathieu ) crystal connected to a non - sinusoidal potential by a transparent domain wall.
|
arxiv:1306.0677
|
convolutional neural networks ( cnns ) have become the dominant neural network architecture for solving visual processing tasks. one of the major obstacles hindering the ubiquitous use of cnns for inference is their relatively high memory bandwidth requirements, which can be a main energy consumer and throughput bottleneck in hardware accelerators. accordingly, an efficient feature map compression method can result in substantial performance gains. inspired by quantization - aware training approaches, we propose a compression - aware training ( cat ) method that involves training the model in a way that allows better compression of feature maps during inference. our method trains the model to achieve low - entropy feature maps, which enables efficient compression at inference time using classical transform coding methods. cat significantly improves the state - of - the - art results reported for quantization. for example, on resnet - 34 we achieve 73. 1 % accuracy ( 0. 2 % degradation from the baseline ) with an average representation of only 1. 79 bits per value. reference implementation accompanies the paper at https : / / github. com / cat - teams / cat
|
arxiv:1909.11481
|
we investigate the behaviour of the linear - response coefficients, when in - plane electric field ( $ \ mathbf e $ ) or / and temperature gradient ( $ \ nabla _ { \ mathbf r } t $ ) is / are applied on a two - dimensional semimetal harbouring anisotropic dirac cones. the anisotropy is caused by ( 1 ) differing fermi velocities along the two mutually perpendicular momentum axes, and ( 2 ) tilting parameters. using the semiclassical boltzmann formalism, we derive the forms of the response coefficients, in the absence and presence of a nonquantizing magnetic field $ \ mathbf b $. the magnetic field affects the response only when it is oriented perpendicular to the plane of the material, with the resulting expressions computed with the help of the so - called lorentz - force operator, appearing in the linearized boltzmann equation. the solution has to be found in a recursive manner, which produces terms in powers of $ | \ mathbf b | $. we discuss the validity of the mott relation and the wiedemann - franz law for the lorentz - operator - induced parts.
|
arxiv:2412.13978
|
we consider oblique incidence of ( p ) tm - polarized wave on the anisotropic superconducting slab, immersed on a dielectric media, such that its uniaxial ( c ) axis is perpendicular to the surfaces. the below and above plasma frequency transmissivity patterns are studied and several of its properties determined, within the context of the maxwell - london theory. below, the regime is attenuated for any incident angle, and there is a transmissivity maximum, quite pronounced in case of a very high external dielectric constant. above, a propagative regime exists where the superconductor is a natural optical resonator, and we show here that the minimum of the transmission peaks are modulated by an envelope function associated to the brewster condition. we propose this set - up to obtain light with an extremely small transverse wavelength inside the superconductor.
|
arxiv:cond-mat/9706184
|
recent quantum oscillation measurements on underdoped cuprates are shown to be consistent with the predictions of a mean field theory of the 1 / 8 magnetic antiphase stripe order proposed to occur in high - $ t _ c $ cuprates. in particular, for intermediate values of the stripe order parameter, the magneto - transport is found to be dominated by an electron pocket.
|
arxiv:0709.0106
|
we carry our numerical simulations of athermally sheared, bidisperse, frictionless disks in two dimensions. from an appropriately defined velocity correlation function, we determine that there are two diverging length scales, $ \ xi $ and $ \ ell $, as the jamming transition is approached. we analyze our results using a critical scaling ansatz for the correlation function, and argue that the more divergent length $ \ ell $ is a consequence of a dangerous irrelevant scaling variable, and that it is $ \ xi $ which is the correlation length that determines the divergence of the system viscosity as jamming is approached from below in the liquid phase. we find that $ \ xi \ sim ( \ phi _ j - \ phi ) ^ { - \ nu } $ diverges with the critical exponent $ \ nu = 1 $. we provide evidence that $ \ xi $ measures the length scale of fluctuations in the rotation of the particle velocity field, while $ \ ell $ measures the length scale of fluctuations in the divergence of the velocity field.
|
arxiv:2004.09311
|
we investigate the structural and thermodynamic properties of a new class of patchy colloids, referred to as inverse patchy colloids ( ipcs ) in their fluid phase via both theoretical methods and simulations. ipcs are nano - or micro - meter sized particles with differently charged surface regions. we extend conventional integral equation schemes to this particular class of systems : our approach is based on the so - called multi - density ornstein - zernike equation, supplemented by the associative percus - yevick approximation ( apy ). to validate the accuracy of our framework, we compare the obtained results with data extracted from $ npt $ and $ nvt $ monte carlo simulations. in addition, other theoretical approaches are used to calculate the properties of the system : the reference hypernetted - chain ( rhnc ) method and the barker - henderson thermodynamic perturbation theory. both apy and rhnc frameworks provide accurate predictions for the pair distribution functions : apy results are in slightly better agreement with mc data, in particular at lower temperatures where the rhnc solution does not converge.
|
arxiv:1412.4492
|
the recent revelation that there are correlated period derivative and pulse shape changes in pulsars has dramatically changed our understanding of timing noise as well as the relationship between the radio emission and the properties of the magnetosphere as a whole. using gaussian processes we are able to model timing and emission variability using a regression technique that imposes no functional form on the data. we revisit the pulsars first studied by lyne et al. ( 2010 ). we not only confirm the emission and rotational transitions revealed therein, but reveal further transitions and periodicities in 8 years of extended monitoring. we also show that in many of these objects the pulse profile transitions between two well - defined shapes, coincident with changes to the period derivative. with a view to the ska and other telescopes capable of higher cadence we also study the detection limitations of period derivative changes.
|
arxiv:1803.02819
|
clinical notes contain information about patients that goes beyond structured data like lab values and medications. however, clinical notes have been underused relative to structured data, because notes are high - dimensional and sparse. this work develops and evaluates representations of clinical notes using bidirectional transformers ( clinicalbert ). clinicalbert uncovers high - quality relationships between medical concepts as judged by humans. clinicalbert outperforms baselines on 30 - day hospital readmission prediction using both discharge summaries and the first few days of notes in the intensive care unit. code and model parameters are available.
|
arxiv:1904.05342
|
in this paper, we present a novel iterative monte carlo method for approximating the stationary probability of a single state of a positive recurrent markov chain. we utilize the characterization that the stationary probability of a state $ i $ is inversely proportional to the expected return time of a random walk beginning at $ i $. our method obtains an $ \ epsilon $ - multiplicative close estimate with probability greater than $ 1 - \ alpha $ using at most $ \ tilde { o } \ left ( t _ { \ text { mix } } \ ln ( 1 / \ alpha ) / \ pi _ i \ epsilon ^ 2 \ right ) $ simulated random walk steps on the markov chain across all iterations, where $ t _ { \ text { mix } } $ is the standard mixing time and $ \ pi _ i $ is the stationary probability. in addition, the estimate at each iteration is guaranteed to be an upper bound with high probability, and is decreasing in expectation with the iteration count, allowing us to monitor the progress of the algorithm and design effective termination criteria. we propose a termination criteria which guarantees a $ \ epsilon ( 1 + 4 \ ln ( 2 ) t _ { \ text { mix } } ) $ multiplicative error performance for states with stationary probability larger than $ \ delta $, while providing an additive error for states with stationary probability less than $ \ delta \ in ( 0, 1 ) $. the algorithm along with this termination criteria uses at most $ \ tilde { o } \ left ( \ frac { \ ln ( 1 / \ alpha ) } { \ epsilon ^ 2 } \ min \ left ( \ frac { t _ { \ text { mix } } } { \ pi _ i }, \ frac { 1 } { \ epsilon \ delta } \ right ) \ right ) $ simulated random walk steps, which is bounded by a constant with respect to the markov chain. we provide a tight analysis of our algorithm based on a locally weighted variant of the mixing time. our results naturally extend for countably infinite state space markov chains via lyapunov function analysis.
|
arxiv:1312.1986
|
the critical behavior of a model with n - vector complex order parameter and three quartic coupling constants that describes phase transitions in unconventional superconductors, helical magnets, stacked triangular antiferromagnets, superfluid helium - 3, and zero - temperature transitions in fully frustrated josephson - junction arrays is studied within the field - theoretical renormalization - group approach in three dimensions. to obtain qualitatively and quantitatively correct results perturbative expansions for \ beta - functions and critical exponents are calculated up to three - loop order and resummed by means of the generalized pade - borel procedure. fixed - point coordinates, critical exponent values, rg flows, etc. are found for the physically interesting cases n = 2 and n = 3. marginal values of n at which the topology of the flow diagram changes are determined as well. in most cases the systems mentioned are shown to undergo fluctuation - driven first - order phase transitions. continuous transitions are allowed in hexagonal d - wave superconductors, in planar helical magnets ( into sinusoidal linearly - polarized phase ), and in triangular antiferromagnets ( into simple unfrustrated ordered states ) with critical exponents \ gamma = 1. 336, \ nu = 0. 677, \ alpha = - 0. 030, \ beta = 0. 347, \ eta = 0. 026, which are hardly believed to be experimentally distinguishable from those of the 3d xy model. the chiral fixed point of rg equations is found to exist and possess some domain of attraction provided n > 3. thus, magnets with heisenberg ( n = 3 ) and xy - like ( n = 2 ) spins should not demonstrate chiral critical behavior with unusual critical exponents ; they can approach the chiral state only via first - order phase transitions.
|
arxiv:cond-mat/9809306
|
understanding plant growth dynamics is essential for applications in agriculture and plant phenotyping. we present the growth modelling ( gromo ) challenge, which is designed for two primary tasks : ( 1 ) plant age prediction and ( 2 ) leaf count estimation, both essential for crop monitoring and precision agriculture. for this challenge, we introduce gromo25, a dataset with images of four crops : radish, okra, wheat, and mustard. each crop consists of multiple plants ( p1, p2,..., pn ) captured over different days ( d1, d2,..., dm ) and categorized into five levels ( l1, l2, l3, l4, l5 ). each plant is captured from 24 different angles with a 15 - degree gap between images. participants are required to perform both tasks for all four crops with these multiview images. we proposed a multiview vision transformer ( mvvt ) model for the gromo challenge and evaluated the crop - wise performance on gromo25. mvvt reports an average mae of 7. 74 for age prediction and an mae of 5. 52 for leaf count. the gromo challenge aims to advance plant phenotyping research by encouraging innovative solutions for tracking and predicting plant growth. the github repository is publicly available at https : / / github. com / mriglab / gromo - plant - growth - modeling - with - multiview - images.
|
arxiv:2503.06608
|
we present a phase - plane analysis of cosmologies containing a barotropic fluid with equation of state $ p _ \ gamma = ( \ gamma - 1 ) \ rho _ \ gamma $, plus a scalar field $ \ phi $ with an exponential potential $ v \ propto \ exp ( - \ lambda \ kappa \ phi ) $ where $ \ kappa ^ 2 = 8 \ pi g $. in addition to the well - known inflationary solutions for $ \ lambda ^ 2 < 2 $, there exist scaling solutions when $ \ lambda ^ 2 > 3 \ gamma $ in which the scalar field energy density tracks that of the barotropic fluid ( which for example might be radiation or dust ). we show that the scaling solutions are the unique late - time attractors whenever they exist. the fluid - dominated solutions, where $ v ( \ phi ) / \ rho _ \ gamma \ to 0 $ at late times, are always unstable ( except for the cosmological constant case $ \ gamma = 0 $ ). the relative energy density of the fluid and scalar field depends on the steepness of the exponential potential, which is constrained by nucleosynthesis to $ \ lambda ^ 2 > 20 $. we show that standard inflation models are unable to solve this ` relic density ' problem.
|
arxiv:gr-qc/9711068
|
old paper on the abstract scattering theory ( st ) of periodic hamiltonians. updating of the references and correction of some minor non - mathematical misprints by h. c. rosu.
|
arxiv:1212.2931
|
higher repetition - rate optical pulse trains have been desired for various applications such as high - bit - rate optical communication, photonic analogue - to - digital conversion, and multi - photon imaging. generation of multi ghz and higher repetition - rate optical pulse trains directly from mode - locked oscillators is often challenging. as an alternative, harmonic injection locking can be applied for extra - cavity repetition - rate multiplication ( rrm ). here we have investigated the operation conditions and achievable performances of all - fibre, highly tunable harmonic injection locking - based pulse rrm. we show that, with slight tuning of slave laser length, highly tunable rrm is possible from a multiplication factor of 2 to > 100. the resulting maximum smsr is 41 db when multiplied by a factor of two. we further characterize the noise properties of the multiplied signal in terms of phase noise and relative intensity noise. the resulting absolute rms timing jitter of the multiplied signal is in the range of 20 fs to 60 fs ( 10 khz - 1 mhz ) for different multiplication factors. with its high tunability, simple and robust all - fibre implementation, and low excess noise, the demonstrated rrm system may find diverse applications in microwave photonics, optical communications, photonic analogue - to - digital conversion, and clock distribution networks.
|
arxiv:1806.10784
|
we give a unified construction of quantum groups, q - boson algebras and quantized weyl algebras and an action of quantum groups on quantized weyl algebras. this enables us to give a conceptual proof of the semi - simplicity of the category $ \ mathcal { o } ( b _ q ) $ introduced by t. nakashima and the classification of all simple objects in it.
|
arxiv:1004.0171
|
this paper investigates the design and performance of delayed bit - interleaved coded modulation ( dbicm ) with low - density parity - check ( ldpc ) codes. for gray labeled square $ m $ - ary quadrature amplitude modulation ( qam ) constellations, we investigate the optimal delay scheme with the largest spectrum efficiency of dbicm for a fixed maximum number of delayed time slots and a given signal - to - noise ratio. when analyzing the capacity of dbicm, we find two important properties : the capacity improvement due to delayed coded bits being mapped to the real and imaginary parts of the transmitted symbols are independent of each other ; a pair of delay schemes with delayed coded bits having identical bit - channel capacity lead to equivalent dbicm capacity. using these two properties, we efficiently optimize the delay scheme for any uniform gray - qam systems. furthermore, these two properties enable efficient ldpc code designs regarding unequal error protection via bit - channel type classifications. moreover, we use protograph - based extrinsic information transfer charts to jointly optimize degree distributions and channel assignments of ldpc codes and propose a constrained progressive edge growth like algorithm to jointly construct ldpc codes and bit - interleavers for dbicm, taking distinctive bit - channel ' s capacity into account. simulation results demonstrate that the designed ldpc coded dbicm systems significantly outperform ldpc coded bicm systems.
|
arxiv:2103.02105
|
we investigate the occurrence and nature of naked singularities in the szekeres space - times. these space - times represent irrotational dust. they do not have any killing vectors and they are generalisations of the tolman - bondi - lemaitre space - times. it is shown that in these space - times there exist naked singularities that satisfy both the limiting focusing condition and the strong limiting focusing condition. the implications of this result for the cosmic censorship hypothesis are discussed.
|
arxiv:gr-qc/9605033
|
recently, lembert, gilson et al proposed a lucid and systematic approach to obtain bilinear b \ " { a } cklund transformations and lax pairs for constant - coefficient soliton equations based on the use of binary bell polynomials. in this paper, we would like to further develop this method with new applications. we extend this method to systematically investigate complete integrability of nonisospectral and variable - coefficient equations. in addiction, a method is described for deriving infinite conservation laws of nonlinear evolution equations based on the use of binary bell polynomials. all conserved density and flux are given by explicit recursion formulas. by taking variable - coefficient kdv and kp equations as illustrative examples, their bilinear formulism, bilinear b \ " { a } cklund transformations, lax pairs, darboux covariant lax pairs and conservation laws are obtained in a quick and natural manner. in conclusion, though the coefficient functions have influences on a variable - coefficient nonlinear equation, under certain constrains the equation turn out to be also completely integrable, which leads us to a canonical interpretation of their $ n $ - soliton solutions in theory.
|
arxiv:1008.4194
|
let $ \ mathcal { l } $ be the noncrossing partition lattice associated to a finite coxeter group $ w $. in this paper we construct explicit bases for the top homology groups of intervals and rank - selected subposets of $ \ mathcal { l } $. we define a multiplicative structure on the whitney homology of $ \ mathcal { l } $ in terms of the basis, and the resulting algebra has similarities to the orlik - solomon algebra. as an application, we obtain four chain complexes which compute the integral homology of the milnor fibre of the reflection arrangement of $ w $, the milnor fibre of the discriminant of $ w $, the hyperplane complement of $ w $ and the artin group of type $ w $, respectively. we also tabulate some computational results on the integral homology of the milnor fibres.
|
arxiv:2206.05721
|
the equation of motion for a domain wall coupled to gravitational field is derived from the nambu - goto action. the domain wall is treated as a source of gravitational field. the perturbed equation is also obtained with gravitational back reaction on the wall motion taken into account. for general spherically symmetric background case, the equation is expressed in terms of the gauge - invariant variables.
|
arxiv:gr-qc/9802036
|
for odd $ n \ geq 5 $, we establish a short exact sequence about motivic double zeta values $ \ zeta ^ { \ mathfrak { m } } ( r, n - r ) $ with $ r \ geq3 $ odd, $ n - r \ geq2 $. from this we classify all the relations among depth - graded motivic double zeta values $ \ zeta ^ { \ mathfrak { m } } ( r, n - r ) $ with $ r \ geq3 $ odd, $ n - r \ geq2 $. as a corollary, we confirm a conjecture of zagier on the rank of a matrix which concerns relations among multiple zeta values of odd weight.
|
arxiv:1710.02244
|
with the exponential growth of the unmanned aerial vehicle ( uav ) industry and a broad range of applications expected to appear in the coming years, the employment of traditional radar systems is becoming increasingly cumbersome for uav supervision. motivated by this emerging challenge, this paper investigates the feasibility of employing integrated sensing and communication ( isac ) systems implemented over current and future wireless networks to perform this task. we propose a sensing mechanism based on the synchronization signal block ( ssb ) in the fifth - generation ( 5g ) standard that performs sensing in a passive bistatic setting. by assuming planar arrays at the sensing nodes and according to the 5g standard, we consider that the ssb signal is sent in a grid of orthogonal beams that are multiplexed in time, with some of them pointing toward a surveillance region where low - altitude drones can be flying. the cramer - rao bound ( crb ) is derived as the theoretical bound for range and velocity estimation. our results demonstrate the potential of employing ssb signals for uav - like target localization at low snr.
|
arxiv:2504.02641
|
clustering properties and peculiar velocities of halos in large - scale structure carry a wealth of cosmological information over a wide range of scales from linear to nonlinear scales. we use halo catalogs in a suite of high - resolution $ n $ - body simulations to construct mock catalogs of galaxies that resemble the sdss - like luminous early - type galaxies at three redshift bins in the range $ 0. 15 \ le z \ le 0. 7 $. to do this we include 10 nuisance parameters to model variations in halo - galaxy connections for each redshift bin ; the halo occupation distribution, and the spatial and velocity distributions of galaxies in the host halos. we evaluate the fisher information matrix for the redshift - space power spectrum of sdss - like galaxies using different sets of the mock catalogs that are generated from changes in each of model parameters ; cosmological parameters ( $ \ sigma _ 8 $ and $ \ omega _ { \ rm m } $ ), the halo - galaxy connection parameters, and the cosmological distances ( $ d _ { \ rm a } $ and $ h $ parameters at each redshift bin ) for modeling an apparent geometrical distortion of the redshift - space power spectrum ( the alcock - paczynski effect ). we show that combining the monopole and quadrupole power spectra of galaxies allows for accurate estimations of the cosmological parameters and the cosmological distances, even after marginalization over the halo - galaxy parameters, by lifting the parameter degeneracies that are otherwise inevitable if either of the two spectra alone is used. when including the galaxy power spectrum information up to $ k = 0. 3 \, h \, $ mpc $ ^ { - 1 } $, we find about factor of 6 gain in the cosmological information content of ( $ \ sigma _ 8, \ omega _ { \ rm m }, d _ { \ rm a } $ ' s and $ h $ ' s ) compared to $ k = 0. 2 \, h \, $ mpc $ ^ { - 1 } $.
|
arxiv:1907.08515
|
we provide an overview of the results we have attained in the last decade on the identification of quantum structures in cognition and, more specifically, in the formalization and representation of natural concepts. we firstly discuss the quantum foundational reasons that led us to investigate the mechanisms of formation and combination of concepts in human reasoning, starting from the empirically observed deviations from classical logical and probabilistic structures. we then develop our quantum - theoretic perspective in fock space which allows successful modeling of various sets of cognitive experiments collected by different scientists, including ourselves. in addition, we formulate a unified explanatory hypothesis for the presence of quantum structures in cognitive processes, and discuss our recent discovery of further quantum aspects in concept combinations, namely, ' entanglement ' and ' indistinguishability '. we finally illustrate perspectives for future research.
|
arxiv:1503.02994
|
we develop a cohomological method to classify amalgams of groups. we generalize this to simplicial amalgams in any concrete category. we compute the non - commutative 1 - cohomology for several examples of amalgams defined over small simplices.
|
arxiv:1509.04679
|
the fermion condensate ( fc ) is investigated for a ( 2 + 1 ) - dimensional massive fermionic field confined on a truncated cone with an arbitrary planar angle deficit and threaded by a magnetic flux. different combinations of the boundary conditions are imposed on the edges of the cone. they include the bag boundary condition as a special case. by using the generalized abel - plana - type summation formula for the series over the eigenvalues of the radial quantum number, the edge - induced contributions in the fc are explicitly extracted. the fc is an even periodic function of the magnetic flux with the period equal to the flux quantum. depending on the boundary conditions, the condensate can be either positive or negative. for a massless field the fc in the boundary - free conical geometry vanishes and the nonzero contributions are purely edge - induced effects. this provides a mechanism for time - reversal symmetry breaking in the absence of magnetic fields. combining the results for the fields corresponding to two inequivalent irreducible representations of the clifford algebra, the fc is investigated in the parity and time - reversal symmetric fermionic models and applications are discussed for graphitic cones.
|
arxiv:2102.12155
|
to find all extreme points of multimodal functions is called extremum problem, which is a well known difficult issue in optimization fields. applying ant colony optimization ( aco ) to solve this problem is rarely reported. the method of applying aco to solve extremum problem is explored in this paper. experiment shows that the solution error of the method presented in this paper is less than 10 ^ - 8. keywords : extremum problem ; ant colony optimization ( aco )
|
arxiv:0911.3209
|
some recent developments in the phenomenology of the lightest scalar glueball are summarised. tools for determining the gluonic content of a resonance of known mass, width and $ j ^ { pc } $ from its branching fraction in radiative quarkonium decays and production cross section in $ \ gamma \ gamma $ collisions are presented. two $ q \ bar { q } - g $ mixing schemes for $ j ^ { pc } = 0 ^ { + + } $, inspired by the lattice, are shown to lead to similar phenomenology that may be tested at bepc and in $ \ gamma \ gamma $ production at lep2.
|
arxiv:hep-ph/9610426
|
in open - domain question answering ( odqa ), it is essential to discern relevant contexts as evidence and avoid spurious ones among retrieved results. the model architecture that uses concatenated multiple contexts in the decoding phase, i. e., fusion - in - decoder, demonstrates promising performance but generates incorrect outputs from seemingly plausible contexts. to address this problem, we propose the multi - granularity guided fusion - in - decoder ( mgfid ), discerning evidence across multiple levels of granularity. based on multi - task learning, mgfid harmonizes passage re - ranking with sentence classification. it aggregates evident sentences into an anchor vector that instructs the decoder. additionally, it improves decoding efficiency by reusing the results of passage re - ranking for passage pruning. through our experiments, mgfid outperforms existing models on the natural questions ( nq ) and triviaqa ( tqa ) datasets, highlighting the benefits of its multi - granularity solution.
|
arxiv:2404.02581
|
sunward - flowing voids above post - coronal mass ejection ( cme ) flare arcades were first discovered using the soft x - ray telescope ( sxt ) aboard yohkoh and have since been observed with trace ( extreme ultra - violet ( euv ) ), soho / lasco ( white light ), soho / sumer ( euv spectra ), and hinode / xrt ( soft x - rays ( sxr ) ). supra - arcade downflow ( sad ) observations suggest that they are the cross - sections of thin flux tubes retracting from a reconnection site high in the corona. supra - arcade downflowing loops ( sadls ) have also been observed under similar circumstances and are theorized to be sads viewed from a perpendicular angle. previous studies have presented detailed sad observations for a small number of flares. in this paper we present a substantial sads and sadls flare catalog. we have applied semi - automatic detection software to several of these events to detect and track individual downflows thereby providing statistically significant samples of parameters such as velocity, acceleration, area, magnetic flux, shrinkage energy, and reconnection rate. we discuss these measurements, how they were obtained, and potential impact on reconnection models.
|
arxiv:1101.1540
|
in this work we introduce the citrusfarm dataset, a comprehensive multimodal sensory dataset collected by a wheeled mobile robot operating in agricultural fields. the dataset offers stereo rgb images with depth information, as well as monochrome, near - infrared and thermal images, presenting diverse spectral responses crucial for agricultural research. furthermore, it provides a range of navigational sensor data encompassing wheel odometry, lidar, inertial measurement unit ( imu ), and gnss with real - time kinematic ( rtk ) as the centimeter - level positioning ground truth. the dataset comprises seven sequences collected in three fields of citrus trees, featuring various tree species at different growth stages, distinctive planting patterns, as well as varying daylight conditions. it spans a total operation time of 1. 7 hours, covers a distance of 7. 5 km, and constitutes 1. 3 tb of data. we anticipate that this dataset can facilitate the development of autonomous robot systems operating in agricultural tree environments, especially for localization, mapping and crop monitoring tasks. moreover, the rich sensing modalities offered in this dataset can also support research in a range of robotics and computer vision tasks, such as place recognition, scene understanding, object detection and segmentation, and multimodal learning. the dataset, in conjunction with related tools and resources, is made publicly available at https : / / github. com / ucr - robotics / citrus - farm - dataset.
|
arxiv:2309.15332
|
artificially engineered geometric phase optical elements may have tunable photonic functionalities owing to sensitivity to external fields, as is the case for liquid crystals based devices. however, a liquid crystal technology combining high - resolution topological ordering with tunable spectral behavior remains elusive. here, by using a magneto - electric external stimulus, we create robust and efficient self - engineered liquid crystal geometric phase vortex masks with broadly tunable operating wavelength, centimeter - size clear aperture, and high - quality topological ordering.
|
arxiv:1803.05973
|
we analyse our recently - published velocity measurements of alpha cen a ( butler et al. 2004 ). after adjusting the weights on a night - by - night basis in order to optimize the window function to minimize sidelobes, we extract 42 oscillation frequencies with l = 0 to 3 and measure the large and small frequency separations. we give fitted relations to these frequencies that can be compared with theoretical models and conclude that the observed scatter about these fits is due to the finite lifetimes of the oscillation modes. we estimate the mode lifetimes to be 1 - 2 d, substantially shorter than in the sun.
|
arxiv:astro-ph/0406471
|
we show that the energy - momentum tensor of homogeneous fields of arbitrary spin in an expanding universe is always isotropic in average provided the fields remain bounded and evolve rapidly compared to the rate of expansion. an analytic expression for the average equation of state is obtained for lagrangians with generic power - law kinetic and potential terms. as an example we consider the behavior of a spin - two field in the standard fierz - pauli theory of massive gravity. the results can be extended to general space - time geometries for locally inertial observers.
|
arxiv:1311.1402
|
we show that the photoassociation ( pa ) rate of ultracold diatomic molecules in the vicinity of a feshbach resonance could be used to probe variations of the electron - to - proton mass ratio $ \ beta = m _ e / m _ p $, a quantity related to other fundamental constants via the $ \ lambda _ { \ mathrm { qcd } } $ scale. the pa rate exhibits two features near a feshbach resonance, a minimum and a maximum, which are very sensitive to the details of the interactions and the exact mass of the system. the effect and detection threshold are illustrated in the formation rates of ultracold li $ _ 2 $ and lina molecules by performing coupled - channel calculations in an external magnetic field. we find that the pa rate is particularly sensitive near narrow feshbach resonances in heavy molecules, where it might be possible to detect relative variability of $ \ beta $ on the order of $ 10 ^ { - 15 } - 10 ^ { - 16 } $. we also use a two - channel model to derive a proportionality relation between the variation of the pa rate and $ \ beta $ applicable to diatomic molecules.
|
arxiv:1402.0494
|
functional mri ( fmri ) and single - cell transcriptomics are pivotal in alzheimer ' s disease ( ad ) research, each providing unique insights into neural function and molecular mechanisms. however, integrating these complementary modalities remains largely unexplored. here, we introduce scbit, a novel method for enhancing ad prediction by combining fmri with single - nucleus rna ( snrna ). scbit leverages snrna as an auxiliary modality, significantly improving fmri - based prediction models and providing comprehensive interpretability. it employs a sampling strategy to segment snrna data into cell - type - specific gene networks and utilizes a self - explainable graph neural network to extract critical subgraphs. additionally, we use demographic and genetic similarities to pair snrna and fmri data across individuals, enabling robust cross - modal learning. extensive experiments validate scbit ' s effectiveness in revealing intricate brain region - gene associations and enhancing diagnostic prediction accuracy. by advancing brain imaging transcriptomics to the single - cell level, scbit sheds new light on biomarker discovery in ad research. experimental results show that incorporating snrna data into the scbit model significantly boosts accuracy, improving binary classification by 3. 39 % and five - class classification by 26. 59 %. the codes were implemented in python and have been released on github ( https : / / github. com / 77yq77 / scbit ) and zenodo ( https : / / zenodo. org / records / 11599030 ) with detailed instructions.
|
arxiv:2502.02630
|
inks of permanent markers and water - proof cosmetics create elastic thin films upon application on a surface. such adhesive materials are deliberately designed to exhibit water - repellent behavior. therefore, patterns made up of these inks become resistant to moisture and cannot be cleaned by water after drying. however, we show that sufficiently slow dipping of such elastic films, which are adhered to a substrate, into a bath of pure water allows complete removal of the hydrophobic coatings. upon dipping, the air - water interface in the bath forms a contact line on the substrate, which exerts a capillary - induced peeling force at the edge of the hydrophobic thin film. we highlight that this capillary peeling process is more effective at lower velocities of the air - liquid interface and lower viscosities. capillary peeling not only removes such thin films from the substrate but also transfers them flawlessly onto the air - water interface.
|
arxiv:1709.00172
|
we prove that the ls category of the symplectic group $ sp ( n ) $ is bounded above by $ \ binom { n + 1 } { 2 } $. this is achieved by computing the number of critical levels of a height function.
|
arxiv:1202.5570
|
we study the approximate state preparation problem on noisy intermediate - scale quantum ( nisq ) computers by applying a genetic algorithm to generate quantum circuits for state preparation. the algorithm can account for the specific characteristics of the physical machine in the evaluation of circuits, such as the native gate set and qubit connectivity. we use our genetic algorithm to optimize the circuits provided by the low - rank state preparation algorithm introduced by araujo et al., and find substantial improvements to the fidelity in preparing haar random states with a limited number of cnot gates. moreover, we observe that already for a 5 - qubit quantum processor with limited qubit connectivity and significant noise levels ( ibm falcon 5t ), the maximal fidelity for haar random states is achieved by a short approximate state preparation circuit instead of the exact preparation circuit. we also present a theoretical analysis of approximate state preparation circuit complexity to motivate our findings. our genetic algorithm for quantum circuit discovery is freely available at https : / / github. com / beratyenilen / qc - ga.
|
arxiv:2210.06411
|
decomposition of the evidence lower bound ( elbo ) objective of vae used for density estimation revealed the deficiency of vae for representation learning and suggested ways to improve the model. in this paper, we investigate whether we can get similar insights by decomposing the elbo for semi - supervised classification using vae model. specifically, we show that mutual information between input and class labels decreases during maximization of elbo objective. we propose a method to address this issue. we also enforce cluster assumption to aid in classification. experiments on a diverse datasets verify that our method can be used to improve the classification performance of existing vae based semi - supervised models. experiments also show that, this can be achieved without sacrificing the generative power of the model.
|
arxiv:2108.12734
|
the $ k ^ - / k ^ + $ ratio in heavy - ion collisions at gsi energies is studied including the properties of the participating hadrons in hot and dense matter. the determination of the temperature and chemical potential at freeze - out conditions compatible with the ratio $ k ^ - / k ^ + $ is very delicate, and depends on the approach adopted for the antikaon self - energy. three approaches for the $ k ^ - $ self - energy are considered : non - interacting $ k ^ - $, on - shell self - energy and single - particle spectral density. with respect to the on - shell approach, the use of an energy dependent $ \ bar { k } $ spectral density, including both s - and p - wave components of the $ \ bar { k } n $ interaction, lowers considerably the freeze - out temperature and gives rise to the " broad - band equilibration " advocated by brown, rho and song.
|
arxiv:hep-ph/0307009
|
we complete the set of string vertices of non - negative dimension by introducing in a consistent manner those moduli spaces which had previously been excluded. as a consequence we obtain a ` geometrised ' string action taking the simple form $ s = f ( \ b ) $ where ` $ \ b $ ' is the sum of the string vertices. that the action satisfies the b - v master equation follows from the recursion relations for the string vertices which take the form of a ` geometrical ' quantum master equation.
|
arxiv:hep-th/9709126
|
let $ s $ be a riemann surface of type $ ( p, n ) $ with $ 3p - 3 + n > 0 $. let $ \ omega $ be a pseudo - anosov map of $ s $ that is obtained from dehn twists along two families $ \ { a, b \ } $ of simple closed geodesics that fill $ s $. then $ \ omega $ can be realized as an extremal teichm \ " { u } ller mapping on a surface of type $ ( p, n ) $ which is also denoted by $ s $. let $ \ phi $ be the corresponding holomorphic quadratic differential on $ s $. in this paper, we compare the locations of some distinguished points on $ s $ in the $ \ phi $ - flat metric to their locations with respect to the complete hyperbolic metric. more precisely, we show that all possible non - puncture zeros of $ \ phi $ must stay away from all closures of once punctured disk components of $ s \ backslash \ { a, b \ } $, and the closure of each disk component of $ s \ backslash \ { a, b \ } $ contains at most one zero of $ \ phi $. as a consequence of the result, we assert that the number of distinct zeros and poles of $ \ phi $ is less than or equal to the number of components of $ s \ backslash \ { a, b \ } $.
|
arxiv:0708.2371
|
the design of research grants has been hypothesized to be a useful tool for influencing researchers and their science. we test this by conducting two thought experiments in a nationally representative survey of academic researchers. first, we offer participants a hypothetical grant with randomized attributes and ask how the grant would influence their research strategy. longer grants increase researchers ' willingness to take risks, but only among tenured professors, which suggests that job security and grant duration are complements. both longer and larger grants reduce researchers ' focus on speed, which suggests a significant amount of racing in science is in pursuit of resources. but along these and other strategic dimensions, the effect of grant design is small. second, we identify researchers ' indifference between the two grant design parameters and find they are very unwilling to trade off the amount of funding a grant provides in order to extend the duration of the grant $ \ unicode { x2014 } $ money is much more valuable than time. heterogeneity in this preference can be explained with a straightforward model of researchers ' utility. overall, our results suggest that the design of research grants is more relevant to selection effects on the composition of researchers pursuing funding, as opposed to having large treatment effects on the strategies of researchers that receive funding.
|
arxiv:2312.06479
|
in an absorbing or an active host medium characterized by a complex refractive index $ n _ 2 = n _ 2 ' + { \ rm i } n _ 2 ' ' $, our previously developed modified dipole long - wave approximation ( mlwa ) is shown to essentially overly with the exact mie theory results for spherical nanoparticle with radius $ a \ lesssim 25 $ nm ( $ a \ lesssim 20 $ nm ) in the case of ag and au ( al and mg ) nanoparticles. the agreement for au and ag ( al and mg ) nanoparticles, slightly better in the case of au than ag, continues to be acceptable up to $ a \ sim 50 $ nm ( $ a \ sim 40 $ nm ), and can be used, at least qualitatively, up to $ a \ sim 70 $ ~ nm ( $ a \ sim 50 $ nm ) correspondingly. a first order analytic perturbation theory ( pt ) in a normalized extinction coefficient, $ \ bar \ kappa = n _ 2 ' ' / n _ 2 ' $, around a nonabsorbing host is developed within the dipole mlwa and its properties are investigated. it is shown that, in a suitable parameter range, the pt can reliably capture the effect of host absorption or gain on the extinction efficiency of various plasmonic nanoparticles.
|
arxiv:2405.08903
|
we report first results from a series of n - body / gasdynamical simulations designed to study the origin of galaxy morphologies in a cold dark matter - dominated universe. the simulations include star formation and feedback and have numerical resolution sufficiently high to allow for a direct investigation of the morphology of simulated galaxies. we find, in agreement with previous theoretical work, that the presence of the main morphological components of galaxies - - disks, spheroids, bars - - is regulated by the mode of gas accretion and intimately linked to discrete accretion events. in the case we present, disks arise from the smooth deposition of cooled gas at the center of dark halos, spheroids result from the stirring of preexisting disks during mergers, and bars are triggered by tides generated by satellites. this demonstrates that morphology is a transient phenomenon within the lifetime of a galaxy and that the hubble sequence reflects the varied accretion histories of galaxies in hierarchical formation scenarios. in particular, we demonstrate directly that disk / bulge systems can be built and rebuilt by the smooth accretion of gas onto the remnant of a major merger and that the present - day remnants of late dissipative mergers between disks are spheroidal stellar systems with structure resembling that of field ellipticals. the perplexing variety of galaxy morphologies is thus highly suggestive of - - and may actually even demand - - a universe where structures have evolved hierarchically.
|
arxiv:astro-ph/0202466
|
current vision language pretraining models are dominated by methods using region visual features extracted from object detectors. given their good performance, the extract - then - process pipeline significantly restricts the inference speed and therefore limits their real - world use cases. however, training vision language models from raw image pixels is difficult, as the raw image pixels give much less prior knowledge than region features. in this paper, we systematically study how to leverage auxiliary visual pretraining tasks to help training end - to - end vision language models. we introduce three types of visual losses that enable much faster convergence and better finetuning accuracy. compared with region feature models, our end - to - end models could achieve similar or better performance on downstream tasks and run more than 10 times faster during inference. compared with other end - to - end models, our proposed method could achieve similar or better performance when pretrained for only 10 % of the pretraining gpu hours.
|
arxiv:2301.07236
|
conventional visualization tools such as phase diagrams and convex hulls are ill - suited to visualize multiple principal element alloys ( mpeas ) due to their large compositional space that cannot be easily projected onto two dimensions. here, symplex plots are introduced to enable the visualization of various properties along special paths in high - dimensional phase spaces of mpeas. these are polar heatmaps that plot properties along high - symmetry paths radiating from the parent equimolar mpea to a set of chosen lower - order compositions. symplex plots capture the changes in the energy landscape along the special paths and help visualize the effect of addition or substitution of components on the alloy stability, which can be especially useful to assess processing pathways for additive manufacturing. thus, symplex plots can help guide design of mpeas by showing connections between compositions and their properties in the high - dimensional phase space with more information concentrated near the equimolar region.
|
arxiv:2504.03973
|
we revisit a recent discussion about the boundary condition at the origin in the schroedinger radial equation for central potentials. using a slight modification of the usual spherical coordinates, the origin of a previously reported dirac delta function term at the origin is clearly shown up. as a consequence, a vanishing boundary condition at the origin must be imposed in solving the radial equation, regardless the kind of potential.
|
arxiv:1305.2782
|
we investigate the time neutral formulation of quantum cosmology of gell - mann and hartle. in particular we study the proposal discussed by them that our universe corresponds to the time symmetric decoherence functional with initial and final density matrix of low entropy. we show that our universe does not correspond to this proposal by investigating the behaviour of small inhomogeneous perturbations around a friedman - robertson - walker model. these perturbations cannot be time symmetric if they were small at the big bang.
|
arxiv:gr-qc/9301005
|
in this work, we consider the particle - hole random phase approximation ( phrpa ), an approximation to the correlation energy in electronic structure, and show that the phrpa energy of the h2 molecule correctly dissociates. that is, as the hydrogen atoms are pulled apart, the phrpa energy of the system converges to twice the phrpa energy of a single hydrogen atom. despite the simplicity of the h2 system, the correct dissociation of h2 is known to be a difficult problem for density functional approximations. as part of our result, we prove that the phrpa correlation energy is well - defined.
|
arxiv:2210.04593
|
simulated annealing - based ontology matching ( sanom ) participates for the second time at the ontology alignment evaluation initiative ( oaei ) 2019. this paper contains the configuration of sanom and its results on the anatomy and conference tracks. in comparison to the oaei 2017, sanom has improved significantly, and its results are competitive with the state - of - the - art systems. in particular, sanom has the highest recall rate among the participated systems in the conference track, and is competitive with aml, the best performing system, in terms of f - measure. sanom is also competitive with logmap on the anatomy track, which is the best performing system in this track with no usage of particular biomedical background knowledge. sanom has been adapted to the hobbit platfrom and is now available for the registered users.
|
arxiv:2006.05219
|
dynamically integrating new or rapidly evolving information after ( large ) language model pre - training remains challenging, particularly in low - data scenarios or when dealing with private and specialized documents. in - context learning and retrieval - augmented generation ( rag ) face limitations, including their high inference costs and their inability to capture global document information. in this paper, we propose a way of modularizing knowledge by training document - level knowledge modules ( kms ). kms are lightweight components implemented as parameter - efficient lora modules, which are trained to store information about new documents and can be easily plugged into models on demand. we show that next - token prediction performs poorly as the training objective for kms. we instead propose deep context distillation : we learn kms parameters such as to simulate hidden states and logits of a teacher that takes the document in context. our method outperforms standard next - token prediction and pre - instruction training techniques, across two datasets. finally, we highlight synergies between kms and rag.
|
arxiv:2503.08727
|
tate introduced in [ ta71 ] the notion of tate algebras to serve, in the context of analytic geometry over the - adics, as a counterpart of polynomial algebras in classical algebraic geometry. in [ cvv19, cvv20 ] the formalism of gr { \ " o } bner bases over tate algebras has been introduced and advanced signature - based algorithms have been proposed. in the present article, we extend the fglm algorithm of [ fglm93 ] to tate algebras. beyond allowing for fast change of ordering, this strategy has two other important benefits. first, it provides an efficient algorithm for changing the radii of convergence which, in particular, makes effective the bridge between the polynomial setting and the tate setting and may help in speeding up the computation of gr { \ " o } bner basis over tate algebras. second, it gives the foundations for designing a fast algorithm for interreduction, which could serve as basic primitive in our previous algorithms and accelerate them significantly.
|
arxiv:2102.05324
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.