{"id":"W7124289677","doi":"10.65109/ldoy7418","title":"Leveraging Sub-Optimal Data for Human-in-the-Loop Reinforcement Learning","year":2024,"lang":"","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Function (biology); Data collection; Human-in-the-loop; Temporal difference learning; Work (physics)","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","scholarly_communication","open_science"],"consensus_categories":[],"category_scores_codex":[0.005508133,0.0006564023,0.0005088851,0.000533414,0.001045155,0.004526624,0.006578317,0.0002073642,0.0002806684],"category_scores_gemma":[0.0003787672,0.0005566452,0.0002406134,0.001344648,0.0001787321,0.002375354,0.003010036,0.001533758,0.0003907784],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002824341,"about_ca_system_score_gemma":0.0004472566,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001200938,"about_ca_topic_score_gemma":0.00001228311,"domain_scores_codex":[0.9934196,0.0003130046,0.001462188,0.001890047,0.001354915,0.001560178],"domain_scores_gemma":[0.995023,0.001200804,0.0002554472,0.003214493,0.0001449304,0.0001613565],"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":[0.00001309253,0.00003049724,0.00009491605,0.0004292866,0.0001117064,0.00009237685,0.004474193,0.9277676,0.0003047112,0.04598135,0.006966474,0.01373381],"study_design_scores_gemma":[0.0006522925,0.0005355353,0.00004384608,0.0005029039,0.00009037543,0.00003977345,0.0007466826,0.9024022,0.0002738212,0.0001417887,0.0939263,0.0006445126],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002041674,0.001000104,0.9806329,0.002611799,0.002088425,0.00147654,0.000002783711,0.0003868006,0.009758984],"genre_scores_gemma":[0.9540868,0.0002259819,0.01389936,0.000821293,0.0006564321,0.00008154121,0.0002465471,0.00008371013,0.02989832],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9667335,"threshold_uncertainty_score":0.9996885,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09075955790988156,"score_gpt":0.3330438801573506,"score_spread":0.242284322247469,"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."}}