{"id":"W3034360859","doi":"","title":"Learning with Good Feature Representations in Bandits and in RL with a Generative Model","year":2020,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Data Stream Mining Techniques","field":"Computer Science","cited_by":57,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Feature (linguistics); Generative model; Generative grammar; Computer science; Artificial intelligence; Feature learning; Machine learning; Linguistics","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":[0.0001396669,0.0001553088,0.0001533308,0.0001963088,0.00006383585,0.0002301274,0.000452266,0.00004346145,0.0000147455],"category_scores_gemma":[0.0001466169,0.0001272661,0.00001156448,0.0002871142,0.00004081209,0.000503702,0.0001968386,0.0007994889,0.000003871478],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004215606,"about_ca_system_score_gemma":0.00008214632,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001090176,"about_ca_topic_score_gemma":0.000352205,"domain_scores_codex":[0.9987976,0.0001169535,0.0001422503,0.0004875796,0.0002999755,0.0001556275],"domain_scores_gemma":[0.9995325,0.00008081152,0.0001085777,0.0001276811,0.00008617042,0.00006426317],"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":[0.0004541698,0.0001201021,0.3867161,0.00001929194,0.00006406941,0.0003392762,0.01305251,0.3971556,0.00232004,0.1753837,0.0001294842,0.0242456],"study_design_scores_gemma":[0.0006227838,0.000390454,0.009510461,0.0001282283,0.000002106011,0.00001787546,0.0001655002,0.9878534,0.0003470824,0.0006155302,0.0001831125,0.0001634667],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.305977,0.00003492242,0.6438507,0.02425132,0.00003225096,0.0003419724,0.00001930018,0.0003920832,0.02510046],"genre_scores_gemma":[0.9239329,0.00002264084,0.07527363,0.00028884,0.00001875973,0.00003010893,0.00004868552,0.00001110798,0.0003733231],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6179559,"threshold_uncertainty_score":0.5189759,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03676985648599293,"score_gpt":0.3016793059080322,"score_spread":0.2649094494220393,"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."}}