{"id":"W2991005916","doi":"10.1109/icst46399.2020.00019","title":"Comparing Offline and Online Testing of Deep Neural Networks: An Autonomous Car Case Study","year":2020,"lang":"en","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":53,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Context (archaeology); Complement (music); Online and offline; Deep neural networks; Statistical hypothesis testing; Artificial neural network; Test data","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":[0.0002509895,0.0001590303,0.0003050059,0.00004649191,0.000147453,0.00008471993,0.0004751755,0.00003288698,0.000004189124],"category_scores_gemma":[0.0002339642,0.0001477371,0.00002286493,0.0004373974,0.00004288025,0.0004458673,0.0007474198,0.000298825,4.048486e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000144564,"about_ca_system_score_gemma":0.00001751474,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009490375,"about_ca_topic_score_gemma":0.0003913972,"domain_scores_codex":[0.9986686,0.0001610326,0.000353399,0.0004365503,0.0001546428,0.0002257704],"domain_scores_gemma":[0.9990346,0.0002397434,0.0001485315,0.0003162292,0.00008146766,0.0001794119],"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":[0.000005283591,0.0001013905,0.154334,0.000009649953,0.00001252047,0.0008331411,0.00271268,0.809073,0.00002938717,0.0001687087,0.000001025386,0.03271926],"study_design_scores_gemma":[0.0005307843,0.0004556684,0.007302289,0.000003549683,0.0000168653,0.0003981474,0.002000487,0.989125,0.000005402515,0.000009255421,0.000002652175,0.0001499594],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.539856,0.00002474455,0.4596288,0.0001059082,0.00005892358,0.0001145611,2.034302e-7,0.000159757,0.00005111603],"genre_scores_gemma":[0.856577,1.695749e-7,0.1430629,0.0001836125,0.0001592195,0.000001473268,0.000001522176,0.00001170951,0.000002435221],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.316721,"threshold_uncertainty_score":0.6024545,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06495997125813802,"score_gpt":0.299791241828165,"score_spread":0.234831270570027,"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."}}