{"id":"W4416159177","doi":"10.48550/arxiv.2511.07272","title":"Revisiting the Neural Tangent Kernel: the role of large width and depth","year":2025,"lang":"","type":"preprint","venue":"ArXiv.org","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Generalization; Artificial neural network; Limit (mathematics); Kernel (algebra); Limiting; Tangent; Property (philosophy)","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.00159318,0.0005256947,0.0005733083,0.0001579336,0.0006497407,0.0002635423,0.002848634,0.0002294567,0.00003470944],"category_scores_gemma":[0.0005151891,0.0003447093,0.0002618003,0.0007496704,0.0004284508,0.0002050893,0.005192686,0.0009318143,0.000008187541],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007620766,"about_ca_system_score_gemma":0.0001784297,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000988746,"about_ca_topic_score_gemma":0.000007403687,"domain_scores_codex":[0.9962648,0.0004717909,0.0009789312,0.001094362,0.000569529,0.0006206268],"domain_scores_gemma":[0.9959916,0.0007569376,0.0009422889,0.001851859,0.0003424709,0.0001148017],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005084603,0.0002000241,0.606403,0.0005518309,0.0004495937,0.00001185159,0.01174181,0.008277514,0.0004671228,0.169392,0.0003000563,0.2021543],"study_design_scores_gemma":[0.0004318695,0.0001081728,0.1740253,0.001176682,0.0002697682,0.00003557021,0.0006108837,0.8037668,0.004428735,0.01064259,0.00383148,0.0006721815],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1764804,0.01226035,0.7928387,0.009706245,0.00130047,0.002777296,0.0000661627,0.0003960056,0.004174418],"genre_scores_gemma":[0.9895087,0.000786176,0.008015329,0.0009636683,0.000271441,0.0001377252,0.000006903933,0.00002597201,0.0002840903],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8130283,"threshold_uncertainty_score":0.9999005,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02769399829919271,"score_gpt":0.2787694954231988,"score_spread":0.2510754971240061,"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."}}