{"id":"W4411162645","doi":"10.1177/03611981251333340","title":"Deep Reinforcement Learning Freeway Controller Chooses Ramp Metering Over Variable Speed Limits","year":2025,"lang":"en","type":"article","venue":"Transportation Research Record Journal of the Transportation Research Board","topic":"Traffic control and management","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Metering mode; Reinforcement; Reinforcement learning; Variable (mathematics); Control theory (sociology); Controller (irrigation); Computer science; Speed limit; Automotive engineering; Artificial intelligence; Engineering; Control (management); Mathematics; Transport engineering; Structural engineering; Mechanical engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.0005612837,0.0005238663,0.0003214329,0.0002268867,0.0002179761,0.0005343182,0.000883918,0.0004468241,0.00134261],"category_scores_gemma":[0.00192245,0.0001621007,0.0002025272,0.000156049,0.0005155288,0.0004234317,0.0003082549,0.0006148891,0.000202162],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001235318,"about_ca_system_score_gemma":0.001305878,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03648403,"about_ca_topic_score_gemma":0.03776146,"domain_scores_codex":[0.9998094,0.00003687062,0.000006255652,0.00005752997,0.0000273206,0.00006266452],"domain_scores_gemma":[0.9994839,0.0002036933,0.00007216051,0.00004243295,0.0001461312,0.00005160114],"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":[0.0001778538,0.0001213463,0.006817391,0.00003691472,0.00002901725,0.00008088285,0.00004696012,0.9593635,0.00278162,0.001712276,0.001073188,0.02775912],"study_design_scores_gemma":[0.00001629778,0.00004423134,0.0008822558,0.000003338681,0.0000068146,0.000006386942,0.0000142786,0.9971956,0.001217669,0.000366945,0.0002419274,0.000004127308],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7925805,0.000163295,0.1917491,0.0004186679,0.0000728597,0.0001487112,0.0002229713,0.001689715,0.01295416],"genre_scores_gemma":[0.9943423,0.00001346253,0.004807479,0.00002359298,0.000002073816,0.00001323022,0.00004248222,0.000007348966,0.0007480288],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03648403,"threshold_uncertainty_score":0.07254332,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03513325150359926,"score_gpt":0.3151262573198766,"score_spread":0.2799930058162774,"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."}}