{"id":"W3180119292","doi":"10.1007/s10664-021-09982-4","title":"Can Offline Testing of Deep Neural Networks Replace Their Online Testing?: A Case Study of Automated Driving Systems.","year":2021,"lang":"en","type":"article","venue":"PubMed","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"H2020 European Research Council; Canada Research Chairs; Ministry of Education; National Research Foundation of Korea; Fonds National de la Recherche Luxembourg; National Research Foundation; Natural Sciences and Engineering Research Council of Canada; European Commission","keywords":"Computer science; Test strategy; Online and offline; Manual testing; Orthogonal array testing; Context (archaeology); White-box testing; Model-based testing; Non-regression testing; Machine learning; Software performance testing; Black-box testing; Artificial intelligence; Test case; Operating system; Software","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003510266,0.0006996022,0.0002759431,0.0004299882,0.0003311274,0.0005427483,0.00137247,0.001347146,0.001558687],"category_scores_gemma":[0.02229806,0.0002032403,0.0002759447,0.0002862667,0.001107168,0.001658965,0.0006986533,0.001185532,0.000397186],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009120153,"about_ca_system_score_gemma":0.000715153,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004765221,"about_ca_topic_score_gemma":0.006336198,"domain_scores_codex":[0.9976421,0.001286511,0.00008915776,0.0003153306,0.0004392154,0.000227643],"domain_scores_gemma":[0.9822676,0.01255647,0.001210368,0.002071891,0.001457195,0.0004364554],"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":[0.003249518,0.003074804,0.123374,0.0008549986,0.0003424067,0.00665243,0.001631621,0.4300817,0.04021594,0.01340866,0.0121753,0.3649387],"study_design_scores_gemma":[0.0001680331,0.001663339,0.0191842,0.0001166935,0.00007488768,0.001298536,0.0007585196,0.9192986,0.03834958,0.01310123,0.005929957,0.00005648314],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.94086,0.0004249785,0.05202699,0.001261599,0.0001049247,0.0001480422,0.0002221912,0.0007765537,0.00417475],"genre_scores_gemma":[0.9916607,0.00003890419,0.00758854,0.0001101078,0.000007400174,0.00002428023,0.0000677822,0.00003342345,0.0004688467],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004765221,"threshold_uncertainty_score":0.01856428,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03895238907324856,"score_gpt":0.2590227423652258,"score_spread":0.2200703532919773,"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."}}