{"id":"W4288048294","doi":"10.1109/dsn-w54100.2022.00020","title":"Towards Building Resilient Ensembles against Training Data Faults","year":2022,"lang":"en","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Construct (python library); Training (meteorology); Training set; Fault injection; Machine learning; Data modeling; Artificial intelligence; Distributed computing; Data mining; Software engineering; Computer network; 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":[],"consensus_categories":[],"category_scores_codex":[0.001395563,0.0001659674,0.0002008111,0.0001484614,0.0007726616,0.0001208238,0.00428508,0.00002983112,0.0001149324],"category_scores_gemma":[0.0004512604,0.0001669223,0.00004614453,0.0005856792,0.00003230209,0.0007211134,0.007667927,0.0004783337,0.00001153669],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001191244,"about_ca_system_score_gemma":0.000196354,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009150758,"about_ca_topic_score_gemma":0.000008777922,"domain_scores_codex":[0.9974964,0.00025737,0.0002821031,0.0008117137,0.0006986882,0.0004536997],"domain_scores_gemma":[0.9978912,0.0002310879,0.000120092,0.001622745,0.00003052579,0.0001043785],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001466076,0.00005608395,0.0002329776,0.00001092738,0.00003483623,0.0001538935,0.004466102,0.4238152,0.001619025,0.1215537,0.004954866,0.4430878],"study_design_scores_gemma":[0.0003376832,0.00004790044,0.0002455965,0.00000900862,0.000005073456,0.00002487289,0.001086319,0.925234,0.00017532,0.001128078,0.07142155,0.0002845681],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02134436,0.00006990352,0.9572444,0.001900746,0.001942171,0.0001227309,0.000009908726,0.0005091179,0.01685667],"genre_scores_gemma":[0.7231205,0.000003522507,0.2757725,0.0006121706,0.0001271702,0.00001160628,0.00002521502,0.00001555815,0.0003117324],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7017761,"threshold_uncertainty_score":0.9557522,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06867818540571483,"score_gpt":0.3158087203122173,"score_spread":0.2471305349065025,"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."}}