{"id":"W4389218123","doi":"10.48550/arxiv.2311.17855","title":"Maximum Entropy Model Correction in Reinforcement Learning","year":2023,"lang":"en","type":"preprint","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Government of Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Principle of maximum entropy; Computer science; Convergence (economics); Bellman equation; Entropy (arrow of time); Algorithm; Mathematical optimization; Function (biology); Applied mathematics; Mathematics; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001699154,0.000837321,0.0008208701,0.001729582,0.0003174828,0.0009110686,0.002738448,0.0009635077,0.00002023356],"category_scores_gemma":[0.0006161855,0.0009712183,0.0003162843,0.001114319,0.00008016788,0.0006284907,0.004505764,0.00358245,0.0001370763],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002364667,"about_ca_system_score_gemma":0.0008435045,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007006759,"about_ca_topic_score_gemma":0.0007627544,"domain_scores_codex":[0.9942796,0.000308604,0.001345384,0.001412543,0.001155013,0.00149885],"domain_scores_gemma":[0.996026,0.0002163035,0.0009087122,0.002288052,0.0002094884,0.0003513815],"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":[0.00002880094,0.00005128474,0.005247109,0.00009383709,0.00004957698,0.0000855842,0.0006549489,0.975008,0.0003527474,0.01189837,0.002866448,0.003663357],"study_design_scores_gemma":[0.0004416239,0.000166945,0.002248281,0.0003662235,0.00002429989,0.00002441093,0.00004699903,0.9870766,0.0008483538,0.007228633,0.0006655133,0.0008620569],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002447549,0.0001773638,0.987433,0.002226352,0.001831871,0.00142564,0.000003424781,0.003433035,0.001021833],"genre_scores_gemma":[0.7316799,0.001111707,0.2404504,0.001139614,0.0002556717,0.002425847,0.0001857592,0.0002472779,0.02250391],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7469826,"threshold_uncertainty_score":0.9996057,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02159192054808646,"score_gpt":0.2504683555440546,"score_spread":0.2288764349959681,"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."}}