{"id":"W4416396310","doi":"10.1016/j.tbs.2025.101171","title":"Analyzing sequential activity and travel decisions with interpretable deep inverse reinforcement learning","year":2025,"lang":"en","type":"article","venue":"Travel Behaviour and Society","topic":"Transportation Planning and Optimization","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"Singapore-MIT Alliance for Research and Technology Centre; University of Florida","keywords":"Interpretability; Reinforcement learning; Travel behavior; Function (biology); Multinomial logistic regression; TRIPS architecture; Artificial neural network; Behavioral modeling; Deep learning","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":[0.000334558,0.00009577366,0.0001380423,0.00004022804,0.0007934141,0.0001053907,0.00005142145,0.00009875236,0.00002895744],"category_scores_gemma":[0.00002168509,0.00008956289,0.00005128082,0.0002087459,0.0001827333,0.0002109376,0.00001099571,0.0001913971,6.13602e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004483847,"about_ca_system_score_gemma":0.0001056568,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001642352,"about_ca_topic_score_gemma":0.001171723,"domain_scores_codex":[0.9992806,0.00005513183,0.0001242957,0.0002008483,0.000153207,0.0001859651],"domain_scores_gemma":[0.9996778,0.00006380979,0.00005323569,0.00005695051,0.00006003392,0.0000882008],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.000162119,0.0001319482,0.6990821,0.00005138509,0.0002353837,0.000009673104,0.257909,0.0179909,0.001794647,0.009951849,0.0004946337,0.01218643],"study_design_scores_gemma":[0.001377036,0.00009273394,0.9051126,0.0001634481,0.0003437111,0.000001473688,0.06402471,0.0278951,0.0003276155,0.0001038868,0.0002265635,0.0003311234],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7439802,0.00003762303,0.2525397,0.0001730085,0.00005145239,0.0001334063,0.000002956318,0.00003989692,0.003041794],"genre_scores_gemma":[0.9939651,0.0003706891,0.002769488,0.00006220827,0.00001290817,0.000006149005,0.00002170865,0.000005499473,0.002786228],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2499849,"threshold_uncertainty_score":0.610238,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01924071360503699,"score_gpt":0.2930527176155078,"score_spread":0.2738120040104708,"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."}}