{"id":"W4206914807","doi":"10.1109/ssci50451.2021.9660081","title":"Investigation of Maximization Bias in Sarsa Variants","year":2021,"lang":"en","type":"article","venue":"2021 IEEE Symposium Series on Computational Intelligence (SSCI)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Maximization; Reinforcement learning; Computer science; Randomness; Artificial intelligence; Q-learning; Expectation–maximization algorithm; Variance (accounting); Machine learning; Statistics; Mathematical optimization; Mathematics; Maximum likelihood; Economics","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":[0.0004619833,0.0002292533,0.0002861233,0.0003249458,0.0001131503,0.0002043458,0.000628616,0.0001148304,0.0001060687],"category_scores_gemma":[0.0002261153,0.0002580828,0.000082859,0.001603536,0.0001363806,0.0009345615,0.000166792,0.0002415231,0.0001212245],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001382831,"about_ca_system_score_gemma":0.0004874345,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002927607,"about_ca_topic_score_gemma":0.0000153108,"domain_scores_codex":[0.997302,0.0002674459,0.0007961372,0.0005696949,0.0007709847,0.0002937331],"domain_scores_gemma":[0.9979909,0.0004517334,0.0003487085,0.0004819289,0.0006290608,0.00009767189],"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":[0.00001547449,0.00004122922,0.001525591,0.00003695125,0.00001664488,0.0000297791,0.000979288,0.9224274,0.002431234,0.07105975,0.00009743626,0.001339207],"study_design_scores_gemma":[0.0001448252,0.0002219528,0.002724304,0.0001698952,0.000006672034,0.00002996502,0.0001172326,0.9136293,0.06408133,0.01847218,0.0001299492,0.0002724339],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01204117,0.00003076722,0.9829642,0.002132302,0.001030439,0.000186708,0.000004739159,0.00004984004,0.001559826],"genre_scores_gemma":[0.8295961,0.0001226706,0.168592,0.0005625127,0.00008557805,0.00001731738,0.0001069778,0.00002228218,0.000894588],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.817555,"threshold_uncertainty_score":0.9999871,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05639518078776527,"score_gpt":0.2753421580849772,"score_spread":0.2189469772972119,"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."}}