{"id":"W4405470007","doi":"10.48550/arxiv.2412.11521","title":"On the Ability of Deep Networks to Learn Symmetries from Data: A Neural Kernel Theory","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute for Catastrophic Loss Reduction","keywords":"Homogeneous space; Artificial neural network; Computer science; Kernel (algebra); Artificial intelligence; Deep neural networks; Cognitive science; Data science; Mathematics; Psychology; Pure mathematics; Geometry","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002986007,0.0007537661,0.0006904603,0.001283463,0.0006485683,0.001614842,0.001406694,0.00166774,0.002148109],"category_scores_gemma":[0.01358952,0.0004539702,0.0009667368,0.0007690533,0.004584317,0.00730123,0.002812876,0.003378062,0.0003940195],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001621166,"about_ca_system_score_gemma":0.0005930586,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002312165,"about_ca_topic_score_gemma":0.001194398,"domain_scores_codex":[0.9992306,0.0002465913,0.00004456671,0.0001935491,0.0001969654,0.00008765339],"domain_scores_gemma":[0.993079,0.004490744,0.000539949,0.001175929,0.0005296794,0.0001846669],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00009078878,0.0000635636,0.002211841,0.00015144,0.00007523654,0.0001304808,0.0002529855,0.1553352,0.003564085,0.7826469,0.002826048,0.05265145],"study_design_scores_gemma":[0.00000853586,0.00003940168,0.0007272962,0.00003508467,0.0000101427,0.00005012661,0.0000260904,0.4689215,0.001008723,0.5283905,0.0007634377,0.00001901565],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06112403,0.001310021,0.924514,0.005054474,0.0001126466,0.00003292744,0.0002041246,0.0003585635,0.00728925],"genre_scores_gemma":[0.9190432,0.00240018,0.07220686,0.0009409025,0.0004338669,0.0001067545,0.0002670835,0.0001691822,0.004432018],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002986007,"threshold_uncertainty_score":0.01579171,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07335424489021844,"score_gpt":0.2080607483630436,"score_spread":0.1347065034728251,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}