{"id":"W3122396129","doi":"10.1016/j.trc.2021.103050","title":"ResLogit: A residual neural network logit model for data-driven choice modelling","year":2021,"lang":"en","type":"preprint","venue":"Transportation Research Part C Emerging Technologies","topic":"Transportation Planning and Optimization","field":"Social Sciences","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Interpretability; Computer science; Multinomial logistic regression; Residual; Logit; Artificial neural network; Artificial intelligence; Discrete choice; Econometrics; Machine learning; Econometric model; Mixed logit; Logistic regression; Algorithm; Mathematics","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":["metaepi_narrow","sts"],"consensus_categories":[],"category_scores_codex":[0.003133165,0.0003958413,0.0005767282,0.0005102237,0.001873798,0.0004600616,0.002217666,0.001093144,0.00003231039],"category_scores_gemma":[0.001032623,0.0004467781,0.0001731485,0.001093083,0.0007871725,0.0006540478,0.0001668035,0.001906332,0.000003239836],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001456615,"about_ca_system_score_gemma":0.001177709,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002953207,"about_ca_topic_score_gemma":0.02368881,"domain_scores_codex":[0.994301,0.000346237,0.0009467013,0.001535505,0.001546983,0.001323527],"domain_scores_gemma":[0.9957392,0.0009170721,0.0003829862,0.001442051,0.001388265,0.0001304241],"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":[0.00006336538,0.00005456656,0.002341901,0.0002144162,0.00008216997,0.00001814171,0.006039964,0.972317,0.000004004708,0.0061206,0.0110971,0.001646763],"study_design_scores_gemma":[0.0003173696,0.00003244893,0.000241555,0.000402418,0.00009323304,9.715024e-8,0.008760378,0.970157,0.00001675941,0.009578681,0.009930126,0.0004699675],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08552036,0.00259358,0.8847208,0.01638928,0.0009902624,0.003086135,0.001606205,0.004636777,0.0004565871],"genre_scores_gemma":[0.8165175,0.005057919,0.1613143,0.00005010893,0.0004801675,0.0009496544,0.01461643,0.00009725293,0.000916647],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7309972,"threshold_uncertainty_score":0.9997984,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3303453341144098,"score_gpt":0.4484011390891473,"score_spread":0.1180558049747375,"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."}}