{"id":"W3128467522","doi":"","title":"Gradient Descent Resists Compositionality","year":2021,"lang":"en","type":"article","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Principle of compositionality; Computer science; Gradient descent; Generalization; Artificial intelligence; Property (philosophy); Artificial neural network; Independence (probability theory); Process (computing); Machine learning; Natural language processing; Theoretical computer science; Programming language; Mathematics; Epistemology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006834955,0.001317606,0.001372429,0.0006824895,0.001079524,0.001154818,0.001231593,0.001454892,0.002786431],"category_scores_gemma":[0.02873556,0.0009055228,0.000848459,0.00039468,0.003426291,0.003067704,0.003826575,0.003004371,0.0007380163],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009654422,"about_ca_system_score_gemma":0.001571898,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001468726,"about_ca_topic_score_gemma":0.0020515,"domain_scores_codex":[0.9969598,0.001162361,0.0001722217,0.0005335486,0.0008964391,0.0002756214],"domain_scores_gemma":[0.9910685,0.004981488,0.0007873124,0.002020363,0.0008702831,0.0002720586],"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":[0.0003629566,0.0001413466,0.00307289,0.0003385335,0.000175393,0.0002878931,0.0004967162,0.4785761,0.0214878,0.3987451,0.00238043,0.09393481],"study_design_scores_gemma":[0.00003126932,0.0001304169,0.0002496173,0.00002932407,0.00001825091,0.00007501923,0.0000309167,0.8321202,0.00507575,0.1604243,0.001801678,0.0000132583],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05828572,0.0001973248,0.932937,0.0006531765,0.00005343394,0.0001053947,0.00002821787,0.0008489323,0.006890834],"genre_scores_gemma":[0.7564841,0.0001811933,0.236029,0.0007794325,0.00006629265,0.000223866,0.0001102058,0.000488758,0.005637103],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006834955,"threshold_uncertainty_score":0.03614712,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0189648016298313,"score_gpt":0.2486587104044005,"score_spread":0.2296939087745692,"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."}}