{"id":"W3211711919","doi":"","title":"On minimizers and convolutional filters: a partial justification for the unreasonable effectiveness of CNNs in categorical sequence analysis.","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Convolutional neural network; Categorical variable; Hash function; Pooling; Sequence (biology); Pattern recognition (psychology); Artificial intelligence; Initialization; Mathematics; Computer science; Gaussian; Algorithm; Statistics; 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":[],"consensus_categories":[],"category_scores_codex":[0.0003875576,0.0001329692,0.0002148778,0.00008805765,0.0000549761,0.00001767343,0.0002090115,0.0001961477,0.000006522756],"category_scores_gemma":[0.0002571097,0.0001243575,0.000142502,0.0002325136,0.000147934,0.000004682434,0.0002062087,0.0001800158,4.315312e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005592849,"about_ca_system_score_gemma":0.0001430632,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001787854,"about_ca_topic_score_gemma":0.00005500305,"domain_scores_codex":[0.9991276,0.0001750906,0.0001475404,0.0003728745,0.00004746469,0.0001294387],"domain_scores_gemma":[0.9990211,0.0003098593,0.0001640372,0.0003550176,0.0001133769,0.00003663617],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000558668,0.00007796165,0.02097761,0.000305528,0.000507779,0.000005434024,0.00008722305,0.9638744,0.002669114,0.01082706,0.00003987484,0.00006932444],"study_design_scores_gemma":[0.001060913,0.0001642786,0.04400335,0.00007691473,0.0007744975,0.00000269372,0.0002836245,0.9480318,0.004155046,0.001024903,0.0001381188,0.0002838126],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8175852,0.00007223692,0.1818614,0.00002832655,0.00006705225,0.000268033,0.00004717826,0.000004575078,0.00006602929],"genre_scores_gemma":[0.9989123,0.0001030121,0.000297681,0.0000203245,0.00001919413,0.000006617647,0.0005378395,0.000006411145,0.00009662955],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1815637,"threshold_uncertainty_score":0.5071152,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05336097870848226,"score_gpt":0.2301072896824181,"score_spread":0.1767463109739358,"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."}}