{"id":"W3203207428","doi":"10.1007/s00521-021-06259-1","title":"Policy invariant explicit shaping: an efficient alternative to reward shaping","year":2021,"lang":"en","type":"article","venue":"Neural Computing and Applications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; University of Alberta; Alberta Machine Intelligence Institute; Compute Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Advice (programming); Computer science; Invariant (physics); Function (biology); Process (computing); Temporal difference learning; Artificial intelligence; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001978673,0.0001544543,0.0001584124,0.0001337711,0.000470981,0.0004058393,0.0006086224,0.00003822299,0.000002295238],"category_scores_gemma":[0.00008907418,0.0001619706,0.00003906481,0.0007912598,0.00002635648,0.0001240822,0.0007072924,0.0001973174,0.00002971816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004129338,"about_ca_system_score_gemma":0.00008192102,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004056359,"about_ca_topic_score_gemma":0.000002807098,"domain_scores_codex":[0.9984667,0.00007794838,0.0002776095,0.0005966807,0.0002458699,0.0003351491],"domain_scores_gemma":[0.998771,0.0001390676,0.000107425,0.0005973551,0.0001567502,0.0002284713],"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":[9.119499e-7,0.00002898829,0.0001267984,0.00001002253,0.000007345182,0.000004889331,0.001129854,0.814015,0.001646338,0.1660343,0.00003241052,0.01696314],"study_design_scores_gemma":[0.0001406401,0.00005453859,0.001527365,0.00003107153,0.000004268059,0.00003970347,0.00009803676,0.9945835,0.001139648,0.0005813637,0.001608425,0.0001914592],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0480654,0.00003524016,0.9464669,0.003547342,0.00008175688,0.0002646926,0.000001139926,0.0002362534,0.001301256],"genre_scores_gemma":[0.9523637,0.000006800762,0.04493324,0.002216783,0.0003280728,0.00002914191,0.000008008827,0.00001332445,0.0001008761],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9042984,"threshold_uncertainty_score":0.6604971,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06130619872746178,"score_gpt":0.3291911299996931,"score_spread":0.2678849312722313,"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."}}