{"id":"W2965050397","doi":"10.24963/ijcai.2019/434","title":"On Principled Entropy Exploration in Policy Optimization","year":2019,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Benchmark (surveying); Entropy (arrow of time); Mathematical optimization; Optimization problem; Convergence (economics); Set (abstract data type); Monotonic function; Constrained optimization problem; Multi-objective optimization; Descent (aeronautics); Machine learning; Algorithm; Mathematics; Economics; Engineering","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003082071,0.001198604,0.001024904,0.0006675164,0.0004809968,0.0008929384,0.0008846472,0.001076384,0.001530293],"category_scores_gemma":[0.01059915,0.0005917735,0.0005661257,0.0004898977,0.002771137,0.001845831,0.002573563,0.001718264,0.0003229481],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007930989,"about_ca_system_score_gemma":0.001339263,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009792205,"about_ca_topic_score_gemma":0.0008775272,"domain_scores_codex":[0.998615,0.0007836652,0.00004867988,0.0001465791,0.0003194215,0.00008667416],"domain_scores_gemma":[0.9954156,0.003744988,0.0002437809,0.0002584586,0.0002189982,0.0001181825],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001098774,0.00005346545,0.0005599983,0.0001595213,0.00005132351,0.00006996744,0.0001362428,0.8764176,0.002395777,0.0840744,0.0007164154,0.03525538],"study_design_scores_gemma":[0.00001699777,0.00007423154,0.00007068957,0.00001988945,0.000006218312,0.00002265318,0.000009112258,0.9612362,0.0005870009,0.03735462,0.0005944028,0.000008117466],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01139989,0.0004931667,0.9845659,0.0003375122,0.00003046995,0.00004119103,0.00001610205,0.0001340569,0.002981659],"genre_scores_gemma":[0.7620971,0.001002123,0.2318785,0.0004341117,0.000157738,0.0004049047,0.00006999146,0.0002058539,0.003749685],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003082071,"threshold_uncertainty_score":0.01629972,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01397620461221087,"score_gpt":0.254103345513721,"score_spread":0.2401271409015101,"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."}}