callensxavier commited on
Commit
e874be8
·
verified ·
1 Parent(s): e109880

Release v1.0: data/code for the hyperbolic conditioning preprint

Browse files
Files changed (5) hide show
  1. README.md +10 -1
  2. conditioning_arb.csv +6 -0
  3. disorder.csv +66 -0
  4. paper.pdf +2 -2
  5. subspace_control.csv +3 -0
README.md CHANGED
@@ -21,6 +21,12 @@ configs:
21
  data_files: exact_rank.csv
22
  - config_name: exact_rank_full_boundary
23
  data_files: exact_rank_full_boundary.csv
 
 
 
 
 
 
24
  - config_name: spectral_gap
25
  data_files: spectral_gap.csv
26
  - config_name: integrator_controls
@@ -44,6 +50,9 @@ inverse conductance problem on hyperbolic lattices* (X. Callens, 2026), included
44
  | `probe_matched.csv` | DtN and Neumann-to-Dirichlet conditioning, full vs subsampled boundary |
45
  | `exact_rank.csv` | Exact Jacobian ranks over GF(p) for two primes; deficiency vs number of unmeasured degree-2 nodes; condition number on the identifiable subspace |
46
  | `exact_rank_full_boundary.csv` | Exact full-boundary Jacobian ranks over GF(p), two primes: full-rank certificates |
 
 
 
47
  | `spectral_gap.csv` | Dirichlet spectral gap, RC relaxation time and stiffness; `exploratory=True` rows were not preregistered |
48
  | `interior_degree.json` | Integer check that interior nodes of the {7,3} truncations have degree 3, and that the interior of G_L is G_(L-1) (coordinates and edge sets) |
49
  | `integrator_controls.csv` | K1 (matrix-exponential known answer) and K2 (steady state = DtN column) controls per integrator; rows with `status` other than `run` were not executed |
@@ -51,7 +60,7 @@ inverse conductance problem on hyperbolic lattices* (X. Callens, 2026), included
51
 
52
  ## Provenance and reproduction
53
 
54
- Generated from commit `a8a5687985e0d51e655e6dc979a0463d1dc08bc0` of
55
  https://github.com/xaviercallens/SocrateAI-Scientific-CondensedMatterTheory, directory
56
  `experiments/track_h_hyperbolic_network`:
57
 
 
21
  data_files: exact_rank.csv
22
  - config_name: exact_rank_full_boundary
23
  data_files: exact_rank_full_boundary.csv
24
+ - config_name: conditioning_arb
25
+ data_files: conditioning_arb.csv
26
+ - config_name: disorder
27
+ data_files: disorder.csv
28
+ - config_name: subspace_control
29
+ data_files: subspace_control.csv
30
  - config_name: spectral_gap
31
  data_files: spectral_gap.csv
32
  - config_name: integrator_controls
 
50
  | `probe_matched.csv` | DtN and Neumann-to-Dirichlet conditioning, full vs subsampled boundary |
51
  | `exact_rank.csv` | Exact Jacobian ranks over GF(p) for two primes; deficiency vs number of unmeasured degree-2 nodes; condition number on the identifiable subspace |
52
  | `exact_rank_full_boundary.csv` | Exact full-boundary Jacobian ranks over GF(p), two primes: full-rank certificates |
53
+ | `conditioning_arb.csv` | (v1.1) log10 condition number in 512-bit ball arithmetic for the float64-singular flat lattices, with certified radii and the two unsaturated controls |
54
+ | `disorder.csv` | (v1.1) log10 kappa (log-parametrised and raw Jacobian) under U[0.5,1.5] and log-uniform [0.1,10] conductances (5 seeds) and x100 / x0.01 defects |
55
+ | `subspace_control.csv` | (v1.1) probe-matched dimensionality control: square lattice sigma_1/sigma_r vs hyperbolic identifiable-subspace kappa |
56
  | `spectral_gap.csv` | Dirichlet spectral gap, RC relaxation time and stiffness; `exploratory=True` rows were not preregistered |
57
  | `interior_degree.json` | Integer check that interior nodes of the {7,3} truncations have degree 3, and that the interior of G_L is G_(L-1) (coordinates and edge sets) |
58
  | `integrator_controls.csv` | K1 (matrix-exponential known answer) and K2 (steady state = DtN column) controls per integrator; rows with `status` other than `run` were not executed |
 
60
 
61
  ## Provenance and reproduction
62
 
63
+ Generated from commit `13b8e69507d1ae61873ae01e8350ed7210f632c4` of
64
  https://github.com/xaviercallens/SocrateAI-Scientific-CondensedMatterTheory, directory
65
  `experiments/track_h_hyperbolic_network`:
66
 
conditioning_arb.csv ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ case,N,E,boundary_size,prec_bits,log10_kappa_arb,max_rel_radius_Ginv,float64_log10_kappa,control_pass
2
+ square R=10 (control),317,592,76,512,9.717820783998446,2.4310848839151277e-119,9.717820796894493,True
3
+ triangular R=6.45 (control),151,401,49,512,8.67244907995092,1.9895185886713393e-120,8.672449080228727,True
4
+ triangular R=10.75,421,1176,84,512,15.994313354282726,2.8909465912264443e-91,,
5
+ square R=16,797,1528,124,512,16.537546103641652,6.787436727649826e-94,,
6
+ triangular R=17.2,1069,3071,133,512,26.94967179994625,2.9526543536146477e-54,,
disorder.csv ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ graph,N,regime,seed,defect_node,defect_depth,log10_kappa_logparam,log10_kappa_raw
2
+ "{7,3} L=2",112,unit,,,,2.462663408217156,2.462663408217156
3
+ "{7,3} L=2",112,"U[0.5,1.5]",0,,,2.7944717056497814,2.6945419427400483
4
+ "{7,3} L=2",112,"U[0.5,1.5]",1,,,2.6796866427776553,2.55293815655171
5
+ "{7,3} L=2",112,"U[0.5,1.5]",2,,,2.6826685078521937,2.558573593354132
6
+ "{7,3} L=2",112,"U[0.5,1.5]",3,,,2.649720472114809,2.4790304575433755
7
+ "{7,3} L=2",112,"U[0.5,1.5]",4,,,2.553242859409693,2.463749909093556
8
+ "{7,3} L=2",112,"logU[0.1,10]",0,,,5.052652322159692,4.399782949409665
9
+ "{7,3} L=2",112,"logU[0.1,10]",1,,,4.343664654447047,4.19046008029365
10
+ "{7,3} L=2",112,"logU[0.1,10]",2,,,4.8038307219357055,4.291309349604336
11
+ "{7,3} L=2",112,"logU[0.1,10]",3,,,4.088434765850973,3.8122443206004073
12
+ "{7,3} L=2",112,"logU[0.1,10]",4,,,3.7728442549723034,3.6744974605653846
13
+ "{7,3} L=2",112,defect x100,,0,3,3.6735442388018034,5.673462766781663
14
+ "{7,3} L=2",112,defect x0.01,,0,3,3.995573342070234,3.933180632803818
15
+ square R=6,113,unit,,,,5.072443594703144,5.072443594703144
16
+ square R=6,113,"U[0.5,1.5]",0,,,5.112561529774751,4.975616917268685
17
+ square R=6,113,"U[0.5,1.5]",1,,,5.436516829143329,5.308891837496105
18
+ square R=6,113,"U[0.5,1.5]",2,,,5.362139923039375,5.237248778686885
19
+ square R=6,113,"U[0.5,1.5]",3,,,5.259884004312309,5.090937745204798
20
+ square R=6,113,"U[0.5,1.5]",4,,,5.05137098590971,4.8864896688948365
21
+ square R=6,113,"logU[0.1,10]",0,,,7.535131288542363,7.099374703513276
22
+ square R=6,113,"logU[0.1,10]",1,,,7.912214545735253,7.664437100661586
23
+ square R=6,113,"logU[0.1,10]",2,,,8.138124883263913,8.075722358964377
24
+ square R=6,113,"logU[0.1,10]",3,,,7.737119455021785,7.479254376209071
25
+ square R=6,113,"logU[0.1,10]",4,,,7.950781699108725,7.426958683161698
26
+ square R=6,113,defect x100,,56,5,6.6001823739444685,8.465066621161405
27
+ square R=6,113,defect x0.01,,56,5,6.650103292427965,4.906927193810162
28
+ triangular R=6.45,151,unit,,,,8.672449080228727,8.672449080228727
29
+ triangular R=6.45,151,"U[0.5,1.5]",0,,,8.752340562297999,8.647379650046187
30
+ triangular R=6.45,151,"U[0.5,1.5]",1,,,8.892446829448351,8.73065643471477
31
+ triangular R=6.45,151,"U[0.5,1.5]",2,,,8.91158232598506,8.74929591509192
32
+ triangular R=6.45,151,"U[0.5,1.5]",3,,,8.703779976909289,8.555239437590163
33
+ triangular R=6.45,151,"U[0.5,1.5]",4,,,8.673981577112736,8.489600318338347
34
+ triangular R=6.45,151,"logU[0.1,10]",0,,,13.091671862910161,12.564315744689358
35
+ triangular R=6.45,151,"logU[0.1,10]",1,,,13.177153435200541,12.618298354765711
36
+ triangular R=6.45,151,"logU[0.1,10]",2,,,13.791254815596364,13.330328150259561
37
+ triangular R=6.45,151,"logU[0.1,10]",3,,,12.179955313332794,11.556361748464271
38
+ triangular R=6.45,151,"logU[0.1,10]",4,,,12.897462510395728,12.296325392486072
39
+ triangular R=6.45,151,defect x100,,63,5,10.386004725079074,12.272461134243942
40
+ triangular R=6.45,151,defect x0.01,,63,5,9.897085278963964,8.206739673248931
41
+ "{7,3} L=3",315,unit,,,,3.2721712304928863,3.2721712304928863
42
+ "{7,3} L=3",315,"U[0.5,1.5]",0,,,3.58269033968672,3.480462876905942
43
+ "{7,3} L=3",315,"U[0.5,1.5]",1,,,3.5161545349845498,3.3903999916557437
44
+ "{7,3} L=3",315,"U[0.5,1.5]",2,,,3.5517646820282507,3.4327150013159375
45
+ "{7,3} L=3",315,"U[0.5,1.5]",3,,,3.4832032074436654,3.332247087129056
46
+ "{7,3} L=3",315,"U[0.5,1.5]",4,,,3.393910755033562,3.3142474610299586
47
+ "{7,3} L=3",315,"logU[0.1,10]",0,,,5.933486357466464,5.276864346868057
48
+ "{7,3} L=3",315,"logU[0.1,10]",1,,,5.19093231871895,5.090121969441008
49
+ "{7,3} L=3",315,"logU[0.1,10]",2,,,5.821503950562347,5.2724834611872815
50
+ "{7,3} L=3",315,"logU[0.1,10]",3,,,5.1307253061899925,4.924151895162715
51
+ "{7,3} L=3",315,"logU[0.1,10]",4,,,4.978140023711665,4.832430208733597
52
+ "{7,3} L=3",315,defect x100,,0,5,4.473321222963652,6.473172398280385
53
+ "{7,3} L=3",315,defect x0.01,,0,5,4.801928269210271,4.715125690989232
54
+ square R=10,317,unit,,,,9.717820796894493,9.717820796894493
55
+ square R=10,317,"U[0.5,1.5]",0,,,9.455209589164825,9.328952598311572
56
+ square R=10,317,"U[0.5,1.5]",1,,,9.569060484404524,9.452456927929383
57
+ square R=10,317,"U[0.5,1.5]",2,,,9.514656559008381,9.352952531906448
58
+ square R=10,317,"U[0.5,1.5]",3,,,9.303660212028245,9.163854088203422
59
+ square R=10,317,"U[0.5,1.5]",4,,,9.48347628838937,9.290220383010508
60
+ square R=10,317,"logU[0.1,10]",0,,,13.82002889466733,13.646860408323983
61
+ square R=10,317,"logU[0.1,10]",1,,,13.732033797138754,13.468146749188817
62
+ square R=10,317,"logU[0.1,10]",2,,,13.134273421773965,12.578537953644094
63
+ square R=10,317,"logU[0.1,10]",3,,,13.453239137267932,13.312041352093097
64
+ square R=10,317,"logU[0.1,10]",4,,,13.260278688178552,12.883455907122402
65
+ square R=10,317,defect x100,,158,9,10.842846147740769,12.699663112229935
66
+ square R=10,317,defect x0.01,,158,9,11.139157422481015,9.438259820263987
paper.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:93cc04b757b920abeae3a90f2b6e8c1e9db30bb03d374a2ffd4e75f242161936
3
- size 390308
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8985e074d251f499d448b0846096908e25d45582c29c8bb89afe18acaf4036d
3
+ size 418729
subspace_control.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ hyperbolic,r,hyp_log10_kappa_identifiable,square,square_E,square_log10_sigma1_over_sigma_r,square_full_log10_kappa,flat_best_subspace_worse_than_hyperbolic
2
+ "{7,3} L=2, 44 probes",117,2.695399727370996,R=6,200,2.1414835622609645,5.072443594703144,False
3
+ "{7,3} L=3, 76 probes",306,2.9699227297327733,R=10,592,3.45123364716429,9.717820796894493,True