{"id":"W3094049233","doi":"10.1609/aaai.v35i13.17371","title":"Amnesiac Machine Learning","year":2021,"lang":"en","type":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Inference; General Data Protection Regulation; Computer security; Inversion (geology); Artificial intelligence; European union; Training set; Machine learning; Data Protection Act 1998; Business","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":[0.003904966,0.0007131664,0.0007009823,0.0007922287,0.0006999276,0.001912596,0.001910956,0.001269062,0.00466656],"category_scores_gemma":[0.02663742,0.0003681022,0.0006800189,0.0006054349,0.003343211,0.00542366,0.003699972,0.00319974,0.001488117],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001119752,"about_ca_system_score_gemma":0.001603351,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001338041,"about_ca_topic_score_gemma":0.001304164,"domain_scores_codex":[0.9953009,0.001460852,0.0003367419,0.0008838919,0.001685457,0.0003322174],"domain_scores_gemma":[0.9794347,0.006317366,0.001847476,0.01075302,0.001290082,0.0003574447],"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":[0.0004888689,0.0003421494,0.01188071,0.0004441896,0.0003321047,0.0008028815,0.001044061,0.04124249,0.01189687,0.2299106,0.0338739,0.6677412],"study_design_scores_gemma":[0.00007888687,0.0003145923,0.004834993,0.000213112,0.0001481013,0.003106683,0.000336068,0.3632919,0.03402848,0.5367463,0.05679928,0.0001015459],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1065771,0.00226132,0.8455113,0.009979825,0.0002861569,0.0002226444,0.0006385227,0.005109336,0.02941384],"genre_scores_gemma":[0.8759817,0.001183797,0.1049396,0.001716528,0.0001810122,0.0001888012,0.0005857986,0.0002438148,0.01497902],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00466656,"threshold_uncertainty_score":0.0206517,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1029753437177614,"score_gpt":0.3104291401143369,"score_spread":0.2074537963965755,"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."}}