{"id":"W3169514089","doi":"","title":"EMaQ: Expected-Max Q-Learning Operator for Simple Yet Effective Offline and Online RL","year":2021,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"Google (Canada); University of Alberta; University of Toronto","funders":"","keywords":"Reinforcement learning; Computer science; Backup; Heuristic; Operator (biology); Online and offline; Offline learning; Simple (philosophy); Artificial intelligence; Generative grammar; Machine learning; Mathematical optimization; Theoretical computer science; Online learning; Mathematics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003741838,0.0002838166,0.0003060751,0.00019077,0.0003211228,0.0005545137,0.0006336739,0.00009746385,0.0002814477],"category_scores_gemma":[0.002162023,0.0002823384,0.00009557708,0.0002334735,0.00005185902,0.000402269,0.0004903233,0.0008313278,0.00003970401],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009253198,"about_ca_system_score_gemma":0.0001279484,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003610417,"about_ca_topic_score_gemma":0.00002570941,"domain_scores_codex":[0.9978267,0.0002624973,0.0003843544,0.0006788222,0.0004998803,0.0003477224],"domain_scores_gemma":[0.997979,0.0006722108,0.0002370268,0.0002739126,0.0007085316,0.0001293354],"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":[0.0001456156,0.0002075016,0.02372496,0.00007704897,0.0003261793,0.000135079,0.001893786,0.6653191,0.01118075,0.2327152,0.0003128591,0.06396199],"study_design_scores_gemma":[0.00112873,0.0004765503,0.002694733,0.0001104027,0.00001249655,0.00003676293,0.0002307774,0.9731099,0.001228838,0.0007876402,0.01986218,0.0003209967],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06487779,0.000131236,0.9254411,0.003099932,0.0006934142,0.0003956419,0.000022545,0.0003496301,0.004988674],"genre_scores_gemma":[0.9710746,0.00008260257,0.02286286,0.0005726183,0.0001969263,0.00005332482,0.0004242998,0.00003416512,0.004698557],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9061968,"threshold_uncertainty_score":0.9999629,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02841059612818234,"score_gpt":0.3196032698915692,"score_spread":0.2911926737633869,"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."}}