{"id":"W4403511313","doi":"10.1109/tse.2024.3482984","title":"TEASMA: A Practical Methodology for Test Adequacy Assessment of Deep Neural Networks","year":2024,"lang":"en","type":"article","venue":"IEEE Transactions on Software Engineering","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Huawei Technologies (Canada); University of Ottawa","funders":"Huawei Technologies; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Science Foundation Ireland","keywords":"Computer science; Artificial neural network; Test (biology); Artificial intelligence; Machine learning; Reliability engineering; Software engineering; Engineering","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":[0.0005591994,0.0002076182,0.0002735881,0.0002473292,0.0001007361,0.000086709,0.0003314867,0.0001332369,0.0000171649],"category_scores_gemma":[0.0003379429,0.0002100004,0.0001841938,0.0005053582,0.00002773204,0.0004551526,0.00000754153,0.0007292984,0.000002056431],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001036469,"about_ca_system_score_gemma":0.00006471713,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005609601,"about_ca_topic_score_gemma":0.000001947684,"domain_scores_codex":[0.9986348,0.00008417285,0.0003193697,0.0004169209,0.000205972,0.0003388],"domain_scores_gemma":[0.9937924,0.005660551,0.00005260898,0.000341661,0.00006442017,0.00008838277],"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":[0.000005360199,0.00004757958,0.00002411523,0.00009228244,0.00005640748,0.00001787658,0.0001127956,0.9420817,0.0003495716,0.003131621,0.000007356603,0.0540733],"study_design_scores_gemma":[0.000204161,0.0001721792,0.0001485849,0.00004615842,0.0000506394,0.00006059052,0.00000901636,0.9982512,0.0005528188,0.0000539789,0.0002624565,0.0001882385],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.000295811,0.00006203168,0.9957079,0.0002414952,0.002620146,0.000243995,0.000007966265,0.00081324,0.00000745375],"genre_scores_gemma":[0.4809994,0.000003038452,0.518791,0.00002085335,0.0000800355,0.0000624723,0.000001287652,0.0000249747,0.00001693245],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4807036,"threshold_uncertainty_score":0.8563566,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03694737320835111,"score_gpt":0.3424103154514501,"score_spread":0.3054629422430989,"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."}}