{"id":"W4319663674","doi":"10.1109/tse.2023.3243522","title":"Black-Box Testing of Deep Neural Networks through Test Case Diversity","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Software Engineering","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":86,"is_retracted":false,"has_abstract":true,"ca_institutions":"École de Technologie Supérieure; University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Black box; Artificial neural network; White-box testing; Test (biology); Diversity (politics); Software testing; Artificial intelligence; Machine learning; Software engineering; Software; Programming language; Software development; Software construction","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001961297,0.0002133772,0.0002174415,0.000194645,0.0004081771,0.00004466133,0.0004964725,0.00009856885,0.000009963758],"category_scores_gemma":[0.0002457732,0.0002420931,0.0001096887,0.00157888,0.00004402843,0.0005398009,0.00004158674,0.0005398779,0.00001483407],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006825513,"about_ca_system_score_gemma":0.00001262416,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001056903,"about_ca_topic_score_gemma":0.000009903293,"domain_scores_codex":[0.9987153,0.00003116373,0.0002487124,0.0003575908,0.0002586013,0.0003886458],"domain_scores_gemma":[0.9978503,0.001480398,0.00007828081,0.0004274286,0.0000800199,0.00008355877],"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":[0.000002302276,0.00002485706,0.0008773172,0.00003240732,0.00002061389,0.0004420955,0.0008086603,0.98279,0.00003653576,0.00003717007,0.000007645405,0.0149204],"study_design_scores_gemma":[0.000226302,0.00006405742,0.0005094325,0.00003359066,0.00002101607,0.0001850446,0.00005083737,0.9981755,0.000488847,0.00001553144,0.000009390241,0.0002204229],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03401556,0.00001235527,0.9632514,0.00002693416,0.0009935932,0.0001136596,0.00000568267,0.001566298,0.00001452481],"genre_scores_gemma":[0.8525928,0.000002893006,0.1472664,0.00002184247,0.00006177479,0.000006616218,8.70671e-7,0.00002642079,0.00002033429],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8185772,"threshold_uncertainty_score":0.9872272,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02156941011280877,"score_gpt":0.236715522135328,"score_spread":0.2151461120225193,"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."}}