{"id":"W3211148793","doi":"10.1109/itsc48978.2021.9564528","title":"Microscopic Model-Based RL Approaches for Traffic Signal Control Generalize Better than Model-Free RL Approaches","year":2021,"lang":"en","type":"article","venue":"","topic":"Traffic control and management","field":"Engineering","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Huawei Technologies","keywords":"Computer science; Reinforcement learning; Intersection (aeronautics); Function (biology); Set (abstract data type); Domain (mathematical analysis); Bellman equation; Network topology; Adaptation (eye); Artificial intelligence; Tree (set theory); Mathematical optimization; Engineering","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":[0.001094114,0.001127571,0.001099961,0.0006412474,0.0002897404,0.0009148198,0.001235347,0.0007847321,0.002657469],"category_scores_gemma":[0.004144816,0.0005744584,0.0006957534,0.0004533076,0.0008778943,0.00210207,0.001224686,0.001732106,0.0004358391],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001059142,"about_ca_system_score_gemma":0.001025987,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006927801,"about_ca_topic_score_gemma":0.007197368,"domain_scores_codex":[0.9994881,0.000147137,0.00003129697,0.0001189495,0.0001681544,0.00004629934],"domain_scores_gemma":[0.998518,0.0008932918,0.000160449,0.0002356686,0.000137527,0.00005507386],"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":[0.000007111325,0.00001133272,0.0001906727,0.00002360874,0.00002115372,0.000007768425,0.00001097636,0.9846437,0.0003786575,0.006407968,0.0002266727,0.008070366],"study_design_scores_gemma":[0.000003108616,0.00000946242,0.00004435309,0.000002280387,0.000003049699,0.000003483879,0.000001392883,0.9953205,0.0001068512,0.004250067,0.0002526974,0.000002670169],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01133887,0.000257879,0.983355,0.0003249017,0.00003902409,0.00002655039,0.00004379193,0.0007748128,0.00383908],"genre_scores_gemma":[0.8692318,0.0006472102,0.1246104,0.0003954834,0.0001222522,0.0001643062,0.0001984215,0.0003558867,0.00427412],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006927801,"threshold_uncertainty_score":0.01377493,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03862313088901556,"score_gpt":0.1971712957064223,"score_spread":0.1585481648174067,"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."}}