{"id":"W2990091474","doi":"10.48550/arxiv.1911.11195","title":"A Novel Unsupervised Post-Processing Calibration Method for DNNS with Robustness to Domain Shift","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"","keywords":"Robustness (evolution); Computer science; Calibration; Artificial intelligence; Range (aeronautics); Scaling; Pattern recognition (psychology); Machine learning; Statistics; Mathematics; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001918102,0.001308575,0.000924232,0.0007671286,0.0005833647,0.0008853231,0.002029202,0.001341198,0.001697673],"category_scores_gemma":[0.005555694,0.0005591915,0.001031977,0.0008109802,0.001272363,0.002001134,0.002542585,0.002713067,0.000687529],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001020756,"about_ca_system_score_gemma":0.001380822,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001836703,"about_ca_topic_score_gemma":0.00278867,"domain_scores_codex":[0.9991518,0.0001721718,0.00005525053,0.0002663965,0.0002870706,0.00006725141],"domain_scores_gemma":[0.9984735,0.0003543376,0.0002511033,0.0003971067,0.0004467355,0.00007718414],"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":[0.0001920222,0.0001134654,0.001843795,0.0001256671,0.0001320618,0.0001180823,0.0002156621,0.5350648,0.03605318,0.01577119,0.003675079,0.406695],"study_design_scores_gemma":[0.000005022182,0.00002808686,0.000221115,0.000008421134,0.000009514062,0.00003337994,0.00000804472,0.9869395,0.008758286,0.003098087,0.0008789657,0.00001156867],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008673914,0.0001363518,0.9895739,0.00009613805,0.00003780882,0.00003230489,0.0000400336,0.0008199415,0.0005897068],"genre_scores_gemma":[0.3627172,0.0002653382,0.6325988,0.0003221173,0.0001258266,0.0001745176,0.0004735531,0.0004644469,0.002858079],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002029202,"threshold_uncertainty_score":0.01014405,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05045522398792803,"score_gpt":0.2274240020226364,"score_spread":0.1769687780347083,"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."}}