{"id":"W2996001434","doi":"10.48550/arxiv.1912.05128","title":"Marginalized State Distribution Entropy Regularization in Policy Optimization","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Regularization (linguistics); Entropy (arrow of time); Mathematical optimization; Econometrics; Mathematics; Economics; Computer science; Statistical physics; Mathematical economics; Political science; Physics; Artificial intelligence; Thermodynamics","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":[0.0002645566,0.0002652678,0.0002814445,0.0004182186,0.00007726478,0.000198005,0.001127854,0.0002411033,0.00001976494],"category_scores_gemma":[0.00009990427,0.0003363983,0.0001038034,0.001099319,0.00005193537,0.0005638506,0.001203211,0.0004481048,0.00006520991],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009293835,"about_ca_system_score_gemma":0.0003863248,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001363312,"about_ca_topic_score_gemma":0.000003492383,"domain_scores_codex":[0.9981794,0.0002301339,0.0002811927,0.0007963624,0.0001478076,0.0003651115],"domain_scores_gemma":[0.9982589,0.00005329711,0.0004191198,0.001005042,0.000177209,0.00008639853],"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":[0.00002408646,0.00002373834,0.002172749,0.00004827087,0.00001971546,0.00002981541,0.00009759761,0.8916856,0.000006794568,0.1057717,0.00005008345,0.00006980251],"study_design_scores_gemma":[0.0007299251,0.00003541882,0.001113102,0.00009098594,0.00001612446,0.000001185802,0.00000872378,0.9928234,0.00004686294,0.004557407,0.0002659422,0.000310987],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004920812,0.00001048997,0.9930552,0.0001701654,0.0003866008,0.0004788626,0.0000128485,0.0001848704,0.0007801583],"genre_scores_gemma":[0.985463,0.0002981299,0.006070187,0.00005684762,0.00004466701,0.000001106263,0.0005517774,0.00001959302,0.007494668],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.986985,"threshold_uncertainty_score":0.9999088,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03583607267643475,"score_gpt":0.1904179922911267,"score_spread":0.1545819196146919,"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."}}