{"id":"W4405778651","doi":"10.1109/iros58592.2024.10802721","title":"Guiding Reinforcement Learning with Incomplete System Dynamics","year":2024,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Honeywell (Canada); University of British Columbia","funders":"","keywords":"Reinforcement learning; Computer science; Dynamics (music); Artificial intelligence; Human–computer interaction; Psychology","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":[],"consensus_categories":[],"category_scores_codex":[0.000431139,0.0001981209,0.0001693148,0.0002042804,0.0002203557,0.0008423779,0.0006981282,0.00004965487,0.00002723025],"category_scores_gemma":[0.00002054184,0.0001501285,0.00005596563,0.0006005722,0.00003074395,0.000623086,0.0003781456,0.0002181266,0.0003332948],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004355278,"about_ca_system_score_gemma":0.00009275546,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004255664,"about_ca_topic_score_gemma":0.000004204845,"domain_scores_codex":[0.998286,0.00004831587,0.0003269595,0.0004118777,0.0005322891,0.0003945589],"domain_scores_gemma":[0.9991691,0.0001180771,0.00007109273,0.0004722048,0.00007309166,0.00009639424],"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":[0.000001281955,9.357323e-7,0.0001393922,0.000097275,0.00002910732,0.00004350726,0.0001707518,0.617368,0.000022486,0.3805158,0.0001468482,0.001464613],"study_design_scores_gemma":[0.0001129709,0.0001909058,0.00002538988,0.0003117951,0.00001095038,0.00008117758,0.0002386696,0.9904788,0.00009506156,0.00002292838,0.008205851,0.000225505],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0002841061,0.00004790878,0.9000255,0.0003107892,0.0004211914,0.0001581381,9.123053e-8,0.00178692,0.09696531],"genre_scores_gemma":[0.9376165,0.000005834396,0.04987761,0.00007348093,0.0000622841,0.00001343665,0.000007746096,0.00002476877,0.01231829],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9373325,"threshold_uncertainty_score":0.8123069,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01837415516122789,"score_gpt":0.2285996606694019,"score_spread":0.210225505508174,"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."}}