{"id":"W4391614448","doi":"10.1145/3643671","title":"Supporting Safety Analysis of Image-processing DNNs through Clustering-based Approaches","year":2024,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada; Fonds National de la Recherche Luxembourg; Université du Luxembourg","keywords":"Computer science; Cluster analysis; Artificial intelligence; Data mining; Software 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001906411,0.001868174,0.0006622476,0.00255647,0.0006219557,0.001162595,0.002233391,0.001349414,0.002252289],"category_scores_gemma":[0.008355197,0.0005545314,0.0009145601,0.0007829618,0.001081982,0.001767645,0.00168954,0.001699929,0.0007485036],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002159444,"about_ca_system_score_gemma":0.001507773,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008583513,"about_ca_topic_score_gemma":0.01198003,"domain_scores_codex":[0.9990427,0.0001895737,0.00006270406,0.0003004587,0.0003013701,0.0001032039],"domain_scores_gemma":[0.9963992,0.001401107,0.0004895933,0.0005678683,0.001059837,0.00008231231],"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":[0.0002765169,0.0001266505,0.003557026,0.0001844783,0.0001163064,0.0002174498,0.000202539,0.7646832,0.0172272,0.005936288,0.002629127,0.2048432],"study_design_scores_gemma":[0.000003344249,0.00001629951,0.0003187592,0.000008353158,0.000006918351,0.00001951585,0.00002066297,0.9875379,0.007281841,0.00448948,0.0002908756,0.000006067572],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04085365,0.0002091077,0.9526072,0.0002011366,0.00003607215,0.0001170398,0.0002251946,0.004165196,0.001585465],"genre_scores_gemma":[0.6034812,0.0001936805,0.3930713,0.0002254841,0.00003620892,0.0001647257,0.0007346067,0.0004704579,0.001622274],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008583513,"threshold_uncertainty_score":0.01706713,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08917448517718657,"score_gpt":0.349973502449047,"score_spread":0.2607990172718604,"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."}}