{"id":"W4313549867","doi":"10.1145/3551349.3556936","title":"SmOOD: Smoothness-based Out-of-Distribution Detection Approach for Surrogate Neural Networks in Aircraft Design","year":2022,"lang":"en","type":"article","venue":"","topic":"Probabilistic and Robust Engineering Design","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Bombardier (Canada); Polytechnique Montréal","funders":"","keywords":"Surrogate model; Computer science; Smoothness; Artificial neural network; Gaussian process; Kriging; Overhead (engineering); Artificial intelligence; Machine learning; Mathematical optimization; Reliability engineering; Gaussian; Engineering; 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":[],"consensus_categories":[],"category_scores_codex":[0.004762302,0.0001573576,0.0003084034,0.0002071164,0.0002037287,0.00005524688,0.0004944746,0.00008048069,0.00007815241],"category_scores_gemma":[0.001201719,0.0001294363,0.0001391878,0.0009954538,0.00006145988,0.0001300697,0.00008122213,0.0002157462,0.000001669121],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001456479,"about_ca_system_score_gemma":0.00007628751,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002980692,"about_ca_topic_score_gemma":0.00001190505,"domain_scores_codex":[0.9975107,0.0003888695,0.0006507225,0.0004630485,0.0006532674,0.0003333947],"domain_scores_gemma":[0.9976119,0.00159215,0.0001747615,0.0003743062,0.000179486,0.00006735419],"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":[0.0001929534,0.0001115166,0.0003395794,0.000007681567,0.000004464431,0.000001218066,0.00003596148,0.9903539,0.0001516998,0.0004397895,0.0006560374,0.007705179],"study_design_scores_gemma":[0.0007561462,0.0002463868,0.0007460574,0.000002115913,0.00001029713,0.000001751534,0.0001653572,0.9950666,0.0005515782,0.001784966,0.0005131723,0.0001555559],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004010343,0.0000627331,0.9942031,0.0000780316,0.0007272085,0.000720993,0.00003455381,0.00008355414,0.00007951985],"genre_scores_gemma":[0.9806151,5.241891e-7,0.01874937,0.00004354257,0.00004216291,0.0003063492,0.00005489761,0.00001494968,0.0001730871],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9766048,"threshold_uncertainty_score":0.5278258,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1095634384970158,"score_gpt":0.3157242924627691,"score_spread":0.2061608539657533,"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."}}