{"id":"W3185323353","doi":"10.1109/deeptest52559.2021.00010","title":"TF-DM: Tool for Studying ML Model Resilience to Data Faults","year":2021,"lang":"en","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Dependability; Computer science; Reliability (semiconductor); Resilience (materials science); Reliability engineering; Data modeling; Focus (optics); Fault (geology); Quality (philosophy); Data mining; Software engineering; Engineering","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":[],"consensus_categories":[],"category_scores_codex":[0.0006980886,0.0001466475,0.0001828921,0.00005901697,0.0002672117,0.0001968612,0.002808436,0.00004697597,0.00002059385],"category_scores_gemma":[0.001638699,0.0001407464,0.00003738873,0.0004853039,0.00001850755,0.001069427,0.003779222,0.0001589369,0.00004056628],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004359206,"about_ca_system_score_gemma":0.0002211517,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001920574,"about_ca_topic_score_gemma":0.00003392501,"domain_scores_codex":[0.9979277,0.00006447306,0.0002551722,0.0009812405,0.0003712067,0.0004002167],"domain_scores_gemma":[0.9970987,0.0004088506,0.00005663219,0.002159766,0.0001697046,0.0001063011],"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":[0.00001356619,0.00005628628,0.0004400449,0.00002038635,0.00001557103,0.00002519288,0.0007883207,0.9029663,0.00115303,0.03914888,0.009348804,0.04602362],"study_design_scores_gemma":[0.0002774209,0.00002636585,0.0001894952,0.00001730526,0.000005948595,0.000004943622,0.00009600818,0.9946507,0.0005044632,0.00153145,0.002484896,0.0002109706],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004096732,0.00002137363,0.990456,0.002346815,0.0006467074,0.0002630063,0.000008948793,0.0002195601,0.001940861],"genre_scores_gemma":[0.3174442,0.000001682923,0.6793792,0.0008880119,0.0000891831,0.00001821183,0.00001118311,0.00001141458,0.002156988],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.3133474,"threshold_uncertainty_score":0.5739473,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07931967226435244,"score_gpt":0.3500755904431777,"score_spread":0.2707559181788252,"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."}}