{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008992288,0.0001368383,0.0001812108,0.0001734002,0.0002483711,0.00008805376,0.0009083666,0.00003345866,0.000004654293],"category_scores_gemma":[0.00005550724,0.0001372357,0.00003292105,0.0003550664,0.00007826695,0.000849132,0.00001811144,0.000071644,4.271842e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007448,"about_ca_system_score_gemma":0.0001449133,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003369397,"about_ca_topic_score_gemma":0.0001392669,"domain_scores_codex":[0.9985588,0.000006354567,0.0003490161,0.0004855756,0.0002745119,0.0003257767],"domain_scores_gemma":[0.9991587,0.00006525913,0.00005748964,0.0005577925,0.0001067629,0.00005396337],"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":[0.00006181901,0.00002105067,0.00001534452,0.0002356313,0.00004132898,2.99135e-7,0.00008851934,0.9823209,0.001855411,0.001428089,0.0003924014,0.01353921],"study_design_scores_gemma":[0.001488829,0.00004678293,0.000740994,0.00002861302,0.000227691,1.857316e-7,0.0001425271,0.9884958,0.00001861588,0.00001136267,0.008657949,0.0001406804],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007516438,0.00004606478,0.9886595,0.00004132282,0.0001780844,0.001207781,0.00202337,0.0001820144,0.0001454241],"genre_scores_gemma":[0.9588681,0.00001074796,0.02453829,0.00002925226,0.00005161477,0.00009589827,0.0163548,0.00001281767,0.00003845345],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9641212,"threshold_uncertainty_score":0.559631,"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."}}