{"id":"W3094334368","doi":"10.1139/cjce-2019-0367","title":"Real-time winter road surface condition monitoring using an improved residual CNN","year":2020,"lang":"en","type":"article","venue":"Canadian Journal of Civil Engineering","topic":"Smart Materials for Construction","field":"Environmental Science","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"Google (Canada); University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Road surface; Residual; Transferability; Transfer of learning; Generalization; Artificial intelligence; Calibration; Convolutional neural network; Snow; Deep learning; Set (abstract data type); Data set; Artificial neural network; Snow removal; Data mining; Machine learning; Computer vision; Meteorology; Engineering; Algorithm; Mathematics; Statistics; Geography","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":true,"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":[0.0001776311,0.0001195626,0.000160347,0.0000582111,0.00007108077,0.00007643694,0.0001602036,0.00005851558,0.0009129504],"category_scores_gemma":[0.00005974971,0.0001332881,0.00004370349,0.0001252599,0.00003848521,0.0006205153,0.00001749085,0.0001388461,0.00003452312],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000373267,"about_ca_system_score_gemma":0.00008993848,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004577735,"about_ca_topic_score_gemma":0.005927816,"domain_scores_codex":[0.9991698,0.00002469493,0.0002784828,0.0001262069,0.0001380129,0.0002628417],"domain_scores_gemma":[0.9991485,0.000009873966,0.0001197545,0.00009217663,0.00001965957,0.000610064],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.000007842657,0.000001882127,0.02035699,0.000007477624,0.00001380768,0.00004942755,0.0003844439,0.1803607,0.798594,0.000001212333,0.0001373823,0.00008476742],"study_design_scores_gemma":[0.002109109,0.000849759,0.449727,0.000503138,0.0002177143,0.001424477,0.0007835969,0.3012723,0.2380563,0.00008048717,0.003372922,0.001603202],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9982185,0.00001600182,0.0006616983,0.00008573374,0.0007526384,0.00005353672,0.000007773289,0.00002113346,0.0001829279],"genre_scores_gemma":[0.9973364,0.000002581075,0.002094971,0.0000144098,0.0005158135,2.941277e-7,0.000001877039,0.00002688226,0.000006777377],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5605377,"threshold_uncertainty_score":0.9996167,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0113074483501035,"score_gpt":0.200834902255646,"score_spread":0.1895274539055425,"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."}}