{"id":"W2994150880","doi":"10.1109/icsme.2019.00078","title":"DeepEvolution: A Search-Based Testing Approach for Deep Neural Networks","year":2019,"lang":"en","type":"article","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Fuzz testing; Computer science; Machine learning; Artificial intelligence; Code coverage; Random testing; Model-based testing; Artificial neural network; Test case; Test strategy; Data mining; Software","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":[0.002311108,0.001297596,0.0008695845,0.001325443,0.0004804387,0.0008509706,0.00261861,0.001549463,0.002491567],"category_scores_gemma":[0.008017203,0.000698704,0.0009406413,0.0004888104,0.001613984,0.001382626,0.001708283,0.001776482,0.0002606768],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001691326,"about_ca_system_score_gemma":0.001376531,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006749374,"about_ca_topic_score_gemma":0.006763929,"domain_scores_codex":[0.9990088,0.0004068487,0.0000608555,0.0001681889,0.000240506,0.0001147444],"domain_scores_gemma":[0.9956384,0.003271645,0.0002818575,0.0002884873,0.0004015312,0.0001180486],"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":[0.00007835346,0.00006869147,0.002130808,0.00007551502,0.00006341688,0.0001223373,0.00007218659,0.9388082,0.003611037,0.007898171,0.0008656647,0.04620571],"study_design_scores_gemma":[0.000006362556,0.00002558016,0.00007014687,0.000006671012,0.000005076558,0.00001196577,0.000005698656,0.9955097,0.0008472862,0.003312184,0.0001965275,0.000002863969],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06611568,0.0004504427,0.9268087,0.000475559,0.00004898323,0.000127399,0.0001551939,0.003157776,0.002660142],"genre_scores_gemma":[0.6829644,0.0001865751,0.3135835,0.0003257944,0.00002834819,0.0003079219,0.0003296531,0.0003888047,0.001884956],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006749374,"threshold_uncertainty_score":0.01342016,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01641999231786189,"score_gpt":0.2397265771014301,"score_spread":0.2233065847835682,"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."}}