{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001456428,0.0009129381,0.001007924,0.000465447,0.0004446363,0.0009405302,0.001766363,0.00140623,0.003975269],"category_scores_gemma":[0.008594183,0.000378768,0.0004498267,0.0004129979,0.001332792,0.001221384,0.001761338,0.002295564,0.000878214],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007462671,"about_ca_system_score_gemma":0.001587469,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001708089,"about_ca_topic_score_gemma":0.001604848,"domain_scores_codex":[0.9987462,0.0003796385,0.00006755534,0.0002456368,0.0003949348,0.000166112],"domain_scores_gemma":[0.9964335,0.001877266,0.0003262297,0.0007624478,0.0003730449,0.0002275279],"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":[0.0005158961,0.0002355171,0.0009544129,0.000158613,0.00005240942,0.0002304529,0.0002017507,0.6425785,0.01889727,0.09218501,0.00385694,0.2401332],"study_design_scores_gemma":[0.0000226968,0.00005443895,0.00005519593,0.00001046818,0.000006895987,0.00002590153,0.000006497336,0.9821879,0.002123867,0.0144028,0.001093891,0.000009298115],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01672667,0.0001892291,0.9767137,0.0002885256,0.00008551867,0.00006913687,0.00003000201,0.001664222,0.004233039],"genre_scores_gemma":[0.790283,0.0001690343,0.2040971,0.0003110849,0.00006846709,0.0001378358,0.00006375252,0.0002757582,0.004593985],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003975269,"threshold_uncertainty_score":0.01329863,"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."}}