{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001708417,0.0004137605,0.0004782463,0.000101305,0.0001022965,0.000130059,0.00038547,0.000160757,0.00004865224],"category_scores_gemma":[0.00001028232,0.0003867358,0.0003172941,0.0001220133,0.00004532745,0.000125522,0.00005479115,0.0001669772,0.000008893656],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007884758,"about_ca_system_score_gemma":0.00007406856,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003776694,"about_ca_topic_score_gemma":0.000141703,"domain_scores_codex":[0.9981728,0.00003026076,0.0003948767,0.0005367914,0.0002260857,0.0006392148],"domain_scores_gemma":[0.99907,0.00007257894,0.00003704503,0.0006302608,0.0000433116,0.0001467597],"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":[0.00004008941,0.0001330704,0.000009458152,0.0002474074,0.0002276521,0.000003913097,0.0001497686,0.971074,0.002157925,0.002772776,0.004602985,0.01858091],"study_design_scores_gemma":[0.004543093,0.00003478639,0.00003555049,0.00001583888,0.0001791307,0.000001307231,0.00006187654,0.9904095,0.00174926,0.0006132901,0.00189301,0.0004633473],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09430522,0.0009986412,0.8994018,0.0014955,0.0001497369,0.0008496286,0.0001060388,0.0006749273,0.002018544],"genre_scores_gemma":[0.9337077,0.000009532568,0.06325333,0.0007333597,0.0001807371,0.000620789,0.0001078213,0.00009357548,0.001293213],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8394024,"threshold_uncertainty_score":0.9998584,"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."}}