{"id":"W4382652027","doi":"10.1101/2023.06.28.546924","title":"Learning better with Dale’s Law: A Spectral Perspective","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Neural dynamics and brain function","field":"Neuroscience","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Recurrent neural network; Singular value decomposition; Computer science; Artificial neural network; Artificial intelligence; Feed forward; Perspective (graphical); Singular value; Constraint (computer-aided design); Law; Mathematics; Engineering; Physics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003228536,0.0006216844,0.0005019318,0.0002796991,0.0004135817,0.0004778235,0.000518623,0.0003460067,0.00004219831],"category_scores_gemma":[0.0003865877,0.0005787087,0.0001741176,0.0006715456,0.0002782543,0.0002455263,0.0005115772,0.00183846,0.0003004009],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004403969,"about_ca_system_score_gemma":0.0002485153,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002444472,"about_ca_topic_score_gemma":0.00002429394,"domain_scores_codex":[0.9962556,0.0002368355,0.0003523597,0.001779953,0.0006149983,0.0007602869],"domain_scores_gemma":[0.9980669,0.0001818141,0.0003559079,0.0008990261,0.0002625286,0.0002338347],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00007929237,0.00008808168,0.002203981,0.000117273,0.00005693734,0.0004473741,0.00002937472,0.0007396138,0.9753519,0.02072121,0.0001642906,6.905341e-7],"study_design_scores_gemma":[0.001481812,0.0008293784,0.08064064,0.0007902434,0.0002471551,3.565175e-7,0.00003834988,0.008156521,0.898815,0.0002324477,0.005762373,0.003005694],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9935467,0.00004057611,0.0006560443,0.00166586,0.001612423,0.0007458621,0.0001180717,0.001427753,0.0001867644],"genre_scores_gemma":[0.9970466,0.0000640982,0.0007078677,0.0009911665,0.0006886405,0.0001487143,2.149384e-7,0.0002397826,0.000112873],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.07843667,"threshold_uncertainty_score":0.9996665,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02141940845334666,"score_gpt":0.2266716778410704,"score_spread":0.2052522693877237,"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."}}