{"id":"W3160165418","doi":"10.1109/icse43902.2021.00044","title":"Self-Checking Deep Neural Networks in Deployment","year":2021,"lang":"en","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"National Satellite of Excellence in Trustworthy Software Systems, National University of Singapore; National Research Foundation Singapore; National University of Singapore; National Research Foundation","keywords":"Software deployment; Computer science; Artificial neural network; Artificial intelligence; Software engineering","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":[0.000245552,0.0001092,0.0001305203,0.00005628245,0.00007529867,0.0001375914,0.0004908498,0.00005439876,0.00004468244],"category_scores_gemma":[0.00006486807,0.0001078581,0.00004329377,0.0005422793,0.000009285788,0.0003521036,0.000618734,0.0003037339,0.00001113198],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007627825,"about_ca_system_score_gemma":0.00003059733,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002797928,"about_ca_topic_score_gemma":0.0001147852,"domain_scores_codex":[0.9987905,0.0001383067,0.0001942338,0.0003654983,0.0001910471,0.0003203918],"domain_scores_gemma":[0.9993705,0.0001170125,0.00004408278,0.0003826282,0.00003805045,0.00004770833],"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":[0.000001503886,0.00004688245,0.01768699,0.00000592317,0.000008934125,0.0002997785,0.0004772549,0.8860254,0.00001713663,0.0493128,0.00002682058,0.04609054],"study_design_scores_gemma":[0.0002460012,0.000008899316,0.003124249,0.00000549995,0.000002169703,0.00002641556,0.00004214313,0.9953181,0.00009421421,0.0004666029,0.0005356251,0.0001301339],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008929478,0.0001906783,0.9843056,0.0008948396,0.0005307749,0.0000574565,1.518255e-8,0.0003242003,0.004766993],"genre_scores_gemma":[0.7925269,0.000006731674,0.206666,0.000608728,0.00008457121,0.000004739264,7.77071e-7,0.000008939882,0.00009262935],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7835975,"threshold_uncertainty_score":0.4398324,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0108506426513521,"score_gpt":0.2484945350835424,"score_spread":0.2376438924321903,"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."}}