{"id":"W4391399751","doi":"10.1613/jair.1.15348","title":"Human-in-the-Loop Reinforcement Learning: A Survey and Position on Requirements, Challenges, and Opportunities","year":2024,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":133,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada); University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; Austrian Science Fund; University of Alberta; Alberta Machine Intelligence Institute; Compute Canada; Canadian Institute for Advanced Research","keywords":"Human-in-the-loop; Reinforcement learning; Position (finance); Loop (graph theory); Computer science; Reinforcement; Artificial intelligence; Business; Psychology; Mathematics; Social psychology","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":[0.003583313,0.0008258613,0.001137092,0.001522648,0.0004460923,0.003093082,0.001996574,0.00208417,0.003353982],"category_scores_gemma":[0.005797259,0.000634996,0.0007480102,0.002350661,0.001521596,0.004334291,0.001376805,0.002571811,0.001341233],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001644091,"about_ca_system_score_gemma":0.001918828,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002632716,"about_ca_topic_score_gemma":0.001618935,"domain_scores_codex":[0.9983942,0.0004878294,0.0001695354,0.0003053081,0.0005459243,0.00009718847],"domain_scores_gemma":[0.9930853,0.005451431,0.0002390788,0.0002432442,0.0008108486,0.0001700731],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00008981721,0.000223206,0.001919179,0.005940475,0.0000780673,0.0001148716,0.0003832243,0.01950579,0.0007945432,0.1032021,0.01055643,0.8571924],"study_design_scores_gemma":[0.00006030313,0.0008401987,0.00303787,0.006986271,0.0001545887,0.0008980132,0.001469581,0.1154659,0.002577621,0.1738071,0.694502,0.0002006193],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.006232893,0.8435751,0.1094961,0.01161769,0.0006909893,0.0001304983,0.0001027697,0.0004066256,0.02774726],"genre_scores_gemma":[0.1110275,0.8267695,0.04985069,0.002618396,0.002094848,0.0002229393,0.0002654457,0.0001619097,0.006988735],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.003583313,"threshold_uncertainty_score":0.01895058,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4886989474982671,"score_gpt":0.4569376480713653,"score_spread":0.03176129942690176,"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."}}