{"id":"W4382318688","doi":"10.1609/aaai.v37i12.26650","title":"On the Challenges of Using Reinforcement Learning in Precision Drug Dosing: Delay and Prolongedness of Action Effects","year":2023,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Receptor Mechanisms and Signaling","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Institut de Valorisation des Données","keywords":"Partially observable Markov decision process; Reinforcement learning; Markov decision process; Computer science; Dosing; Action (physics); Markov chain; Task (project management); Markov process; Artificial intelligence; Baseline (sea); Machine learning; Markov model; Medicine; Pharmacology; Mathematics; Biology","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":[0.002430322,0.0008450405,0.00110735,0.0002508001,0.0003886725,0.0009928647,0.001088791,0.001259199,0.001584503],"category_scores_gemma":[0.007133449,0.0004316276,0.0005643021,0.0003561993,0.001418066,0.001606452,0.001029947,0.00239741,0.000206696],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0010619,"about_ca_system_score_gemma":0.002346541,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006015356,"about_ca_topic_score_gemma":0.004296019,"domain_scores_codex":[0.9991253,0.0003474022,0.00005540514,0.0002219518,0.0001781378,0.00007176949],"domain_scores_gemma":[0.9951716,0.003691366,0.0004559259,0.0002926732,0.0002328484,0.0001555973],"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":[0.000113234,0.00006457629,0.000605193,0.000120346,0.00004620657,0.00008329582,0.00006909556,0.9296833,0.002545981,0.01772143,0.0006341106,0.04831317],"study_design_scores_gemma":[0.00002736613,0.00008376271,0.0001691897,0.00001474253,0.0000131214,0.00003648822,0.00001268217,0.9789391,0.0010175,0.01889988,0.0007750893,0.00001120517],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02614199,0.0007437521,0.9690238,0.00108983,0.00005314908,0.00005335181,0.00004865401,0.0004042056,0.002441366],"genre_scores_gemma":[0.8711298,0.0005371362,0.1258923,0.0004168001,0.00007246009,0.00008783489,0.00005554002,0.00007379607,0.001734306],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006015356,"threshold_uncertainty_score":0.01285291,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07934898163826612,"score_gpt":0.3113257603477342,"score_spread":0.2319767787094681,"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."}}