{"id":"W4381733153","doi":"10.3233/faia230098","title":"Reinforcement Learning Requires Human-in-the-Loop Framing and Approaches","year":2023,"lang":"en","type":"book-chapter","venue":"Frontiers in artificial intelligence and applications","topic":"Mental Health Research Topics","field":"Psychology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Human-in-the-loop; Framing (construction); Computer science; Markov decision process; Set (abstract data type); Artificial intelligence; Markov process; Engineering; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001583862,0.0006255908,0.0004482397,0.0003049681,0.000457433,0.002779635,0.001404691,0.001706536,0.00637092],"category_scores_gemma":[0.002428909,0.000272809,0.000394201,0.0003728235,0.004711838,0.00279282,0.001354137,0.003406622,0.00152016],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002098473,"about_ca_system_score_gemma":0.00115157,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00119287,"about_ca_topic_score_gemma":0.001131752,"domain_scores_codex":[0.9992923,0.0003265901,0.00002468237,0.0001247126,0.0001926969,0.00003898385],"domain_scores_gemma":[0.9983305,0.001253352,0.00006200642,0.000164299,0.0001160509,0.00007377496],"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":[0.000006297491,0.00001753171,0.00005634431,0.00009121717,0.000008325183,0.00002635674,0.0001691554,0.008893808,0.0002088579,0.9549676,0.003780322,0.0317742],"study_design_scores_gemma":[0.000005824576,0.00001600228,0.0000560906,0.00006509385,0.000003948392,0.00003587324,0.00005254481,0.01879617,0.0002470692,0.9289269,0.05178566,0.00000867085],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003476019,0.009804166,0.719906,0.0169022,0.0008269019,0.00007427525,0.0000689575,0.0003398845,0.2486016],"genre_scores_gemma":[0.4705925,0.02141789,0.3731662,0.004952861,0.001858038,0.0005211639,0.0001601102,0.0003568422,0.1269744],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00637092,"threshold_uncertainty_score":0.02131289,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2781771810200347,"score_gpt":0.4138547181675101,"score_spread":0.1356775371474753,"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."}}