{"id":"W7104642538","doi":"10.2139/ssrn.5734559","title":"Deep Reinforcement Learning for Dynamic Origin-Destination Matrix Estimation in Microscopic Traffic Simulations Considering Credit Assignment","year":2025,"lang":"","type":"preprint","venue":"SSRN Electronic Journal","topic":"Traffic control and management","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Process (computing); Calibration; Markov decision process; Matrix (chemical analysis); Markov process; Link (geometry)","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":[0.001337912,0.0006370239,0.001253188,0.0004402013,0.0004618235,0.000924068,0.001414772,0.001615949,0.002203305],"category_scores_gemma":[0.008393957,0.0007125652,0.0004197297,0.0004046386,0.00127934,0.001310698,0.001330978,0.001924731,0.0001983328],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001768354,"about_ca_system_score_gemma":0.001726705,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02327867,"about_ca_topic_score_gemma":0.0132537,"domain_scores_codex":[0.9997075,0.0001164863,0.00001124541,0.00006105624,0.00004087462,0.00006283573],"domain_scores_gemma":[0.995681,0.003375069,0.0002729346,0.0001428035,0.0003126067,0.0002154914],"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":[0.00002047941,0.00001161996,0.0002493333,0.000007501035,0.000005714405,0.000009129546,0.000008292017,0.9964176,0.00008594341,0.001487765,0.00009503276,0.001601571],"study_design_scores_gemma":[0.000001018081,0.000001345194,0.00001368226,4.643164e-7,3.516092e-7,4.274347e-7,6.610194e-7,0.9995387,0.00001523858,0.0004217399,0.000005934946,4.713643e-7],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2924652,0.0004121563,0.7013758,0.001321569,0.0001083417,0.00006106963,0.0001702774,0.0005449465,0.003540518],"genre_scores_gemma":[0.9856327,0.00006026866,0.01271636,0.00006866546,0.00002318278,0.00003764488,0.00007178812,0.00003144711,0.001358112],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02327867,"threshold_uncertainty_score":0.04628628,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007643648704602245,"score_gpt":0.2678890472239066,"score_spread":0.2602453985193043,"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."}}