{"id":"W4415816809","doi":"10.1007/s42421-025-00139-z","title":"A Fully Data-Driven Approach for Realistic Traffic Signal Control Using Offline Reinforcement Learning","year":2025,"lang":"en","type":"article","venue":"Data Science for Transportation","topic":"Traffic control and management","field":"Engineering","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Baidu","keywords":"Reinforcement learning; Construct (python library); Control (management); SIGNAL (programming language); State (computer science); Inference; Traffic flow (computer networking)","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":[0.0008913029,0.0007043435,0.001146406,0.0002957918,0.0003962294,0.0008809016,0.001457886,0.001322825,0.003729716],"category_scores_gemma":[0.003471072,0.0007270295,0.0006259603,0.0002874402,0.0008419341,0.0009821539,0.001396383,0.001927191,0.0004533907],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000856806,"about_ca_system_score_gemma":0.001677379,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009034621,"about_ca_topic_score_gemma":0.007334649,"domain_scores_codex":[0.9996243,0.0001179173,0.00001501114,0.00007465335,0.0001125894,0.00005551719],"domain_scores_gemma":[0.9984919,0.0009463384,0.00009415847,0.0001170709,0.0002516051,0.00009890806],"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":[0.00002803782,0.00002500859,0.0001012148,0.00001772116,0.000009796062,0.00002333597,0.00001450988,0.9891391,0.0004030279,0.004053886,0.0002700636,0.005914248],"study_design_scores_gemma":[0.000002182843,0.000003095565,0.000007525908,7.483656e-7,6.453228e-7,0.000001249125,5.535796e-7,0.9990146,0.00004546189,0.0008817259,0.0000413925,8.3366e-7],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01194536,0.00007242214,0.9851047,0.0001615081,0.00004038129,0.00003557586,0.00005167324,0.0002730684,0.002315311],"genre_scores_gemma":[0.8680462,0.00007040197,0.1267418,0.0001468133,0.00006633412,0.0001747228,0.0001459022,0.0001565844,0.004451152],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009034621,"threshold_uncertainty_score":0.01796407,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04114882550449673,"score_gpt":0.2831969920624237,"score_spread":0.2420481665579269,"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."}}