{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01609841,0.0001747721,0.000245735,0.001040339,0.000399817,0.000978554,0.000853067,0.00009481851,0.00002484346],"category_scores_gemma":[0.0006215171,0.0001332765,0.00006062966,0.0006302557,0.000337882,0.001026968,0.0002422873,0.001157163,0.00003799869],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000164014,"about_ca_system_score_gemma":0.0002109403,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001747567,"about_ca_topic_score_gemma":0.0002119761,"domain_scores_codex":[0.9954813,0.001348796,0.0008657393,0.0003860165,0.001365256,0.0005529216],"domain_scores_gemma":[0.9973912,0.001399598,0.0001405991,0.0003223876,0.0005749958,0.0001712666],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00007418684,0.0001309756,0.00002582832,0.00008239266,0.00002895186,0.0006581799,0.008379174,0.0007639015,0.001902539,0.6098085,0.0002557014,0.3778896],"study_design_scores_gemma":[0.0002084123,0.01844046,0.003094118,0.003374224,0.00004051615,0.001221167,0.06385919,0.2636647,0.06693105,0.5611815,0.01660659,0.001378051],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7477316,0.04249654,0.1380889,0.05591502,0.001609284,0.001782524,0.000004152223,0.0001609901,0.01221098],"genre_scores_gemma":[0.9889548,0.01029009,0.0001776831,0.0001011644,0.0001696334,0.00001189565,0.000001244325,0.00001348956,0.0002799479],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3765116,"threshold_uncertainty_score":0.9436218,"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."}}