{"id":"W2951513593","doi":"10.1016/j.cogpsych.2017.11.002","title":"Compositional inductive biases in function learning","year":2017,"lang":"en","type":"article","venue":"Cognitive Psychology","topic":"Cognitive Science and Education Research","field":"Neuroscience","cited_by":97,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Companhia Brasileira de Metalurgia e Mineração; National Science Foundation","keywords":"Principle of compositionality; Inductive bias; Computer science; Predictability; Artificial intelligence; Property (philosophy); Cognition; Prior probability; Numerosity adaptation effect; Natural language processing; Machine learning; Cognitive psychology; Bayesian probability; Multi-task learning; Psychology; Mathematics; Statistics; Task (project management)","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0003447995,0.00009451213,0.0001103827,0.0002668842,0.0007236807,0.0001133542,0.0002670406,0.00006304709,0.0008293051],"category_scores_gemma":[0.00483997,0.00009496453,0.00003491388,0.0002025817,0.0008284185,0.0004236146,0.00009312585,0.000411912,0.0008755107],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002719626,"about_ca_system_score_gemma":0.00007033768,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002909708,"about_ca_topic_score_gemma":0.00002293451,"domain_scores_codex":[0.9984781,0.0002887125,0.0001355863,0.0005253673,0.0002159059,0.0003563851],"domain_scores_gemma":[0.9988021,0.0006766254,0.0001086154,0.0001556998,0.0001724932,0.0000844698],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0007709741,0.0008567259,0.2817487,0.000006753679,0.00001098007,0.0001108386,0.0009599149,0.000005704367,0.4810338,0.002419167,0.0007582668,0.2313183],"study_design_scores_gemma":[0.0009136755,0.000267644,0.9680234,0.00005266674,0.000004310704,0.00002678559,0.000764902,0.00006560397,0.02510054,0.003449221,0.001192314,0.0001389453],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.901089,0.00001094676,0.0004654481,0.001740161,0.0005228026,0.0001698612,0.000009235579,0.00002328796,0.09596933],"genre_scores_gemma":[0.9972332,0.00003120897,0.00001320068,0.001880224,0.0001348347,0.00006808449,0.000008495847,0.000007597117,0.0006231733],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6862747,"threshold_uncertainty_score":0.9999024,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3245244709197464,"score_gpt":0.5116112390370984,"score_spread":0.187086768117352,"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."}}