{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001347126,0.0004146711,0.0004315427,0.0003955488,0.0004737193,0.0004089169,0.001836361,0.0003135079,0.00001085477],"category_scores_gemma":[0.0005008247,0.0004300707,0.0002318073,0.001287634,0.0000919555,0.0007574417,0.0005859645,0.0008215638,0.000009352417],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004649332,"about_ca_system_score_gemma":0.0002226609,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001143337,"about_ca_topic_score_gemma":0.00008095424,"domain_scores_codex":[0.9965885,0.0002748359,0.0004852258,0.0009231861,0.0005044766,0.001223763],"domain_scores_gemma":[0.997119,0.0006733143,0.0002543568,0.001373994,0.000265811,0.0003134894],"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":[0.00004565508,0.00008837281,0.03552597,0.00004590032,0.00001770502,0.00000914131,0.00007528823,0.9093027,0.0002747938,0.02107434,0.00008118307,0.03345897],"study_design_scores_gemma":[0.0008873181,0.0002400623,0.007781988,0.00002185172,0.00001529793,0.00005465237,0.00003174539,0.98958,0.0001841856,0.0005922978,0.0001415326,0.000469115],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007368577,0.0003451721,0.9871681,0.001174993,0.000283614,0.001525197,0.00000392417,0.001488538,0.0006418356],"genre_scores_gemma":[0.5043102,0.000001098402,0.4942404,0.0008591914,0.0001643619,0.0002719259,0.00001706355,0.00004482461,0.00009093988],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.4969416,"threshold_uncertainty_score":0.9998151,"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."}}