{"id":"W2972204217","doi":"10.1109/icsme.2019.00078","title":"DeepEvolution: A Search-Based Testing Approach for Deep Neural Networks","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Fuzz testing; Computer science; Machine learning; Artificial intelligence; Code coverage; Artificial neural network; Random testing; Deep neural networks; Model-based testing; Training set; Test case; Data mining; Software; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00239942,0.001306934,0.0008857648,0.001328675,0.0004758491,0.0008954953,0.002636236,0.001687355,0.002563825],"category_scores_gemma":[0.008732006,0.0007135331,0.0009778081,0.0005254203,0.001709539,0.001476593,0.001781951,0.00188724,0.0002780784],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001662492,"about_ca_system_score_gemma":0.00131798,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006103616,"about_ca_topic_score_gemma":0.005934011,"domain_scores_codex":[0.998928,0.0004463842,0.00006550128,0.0001767914,0.0002635295,0.0001197575],"domain_scores_gemma":[0.9952902,0.003529087,0.0002924237,0.0003371039,0.0004253729,0.0001258079],"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":[0.00009177344,0.00007674777,0.002279875,0.00008833614,0.0000687752,0.0001349465,0.00008425364,0.9307851,0.004098879,0.009381711,0.0009973134,0.0519124],"study_design_scores_gemma":[0.000007539036,0.00002773,0.00007651346,0.000007731493,0.000005641692,0.00001382884,0.000006454885,0.9944335,0.0009967063,0.004197837,0.0002235144,0.000003102586],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06546822,0.0004699987,0.9272485,0.0004935559,0.00004999487,0.0001248454,0.0001702994,0.003273899,0.002700551],"genre_scores_gemma":[0.6758232,0.0002000376,0.3205097,0.0003403994,0.00003139029,0.0003083036,0.0003644327,0.0004400726,0.001982355],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006103616,"threshold_uncertainty_score":0.01268947,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04957674376864774,"score_gpt":0.2851989741955164,"score_spread":0.2356222304268687,"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."}}