{"id":"W2801491268","doi":"10.1515/popets-2018-0024","title":"Privacy-preserving Machine Learning as a Service","year":2018,"lang":"en","type":"article","venue":"Proceedings on Privacy Enhancing Technologies","topic":"Cryptography and Data Security","field":"Computer Science","cited_by":255,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan","funders":"Center for Discrete Mathematics and Theoretical Computer Science; National Science Foundation","keywords":"Computer science; Homomorphic encryption; Encryption; Cloud computing; Machine learning; Artificial intelligence; Artificial neural network; Deep learning; Domain (mathematical analysis); Information privacy; Service provider; Raw data; Service (business); Data mining; Computer security","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":["metaepi_narrow","open_science"],"consensus_categories":[],"category_scores_codex":[0.0006668227,0.0004030585,0.000341463,0.0006226062,0.0007020821,0.0006087666,0.005592371,0.0002801659,0.00003054336],"category_scores_gemma":[0.003605132,0.0003695879,0.00009878235,0.002469159,0.0001836205,0.001608752,0.00619003,0.0008991635,0.0003131327],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007476731,"about_ca_system_score_gemma":0.00005432676,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009913662,"about_ca_topic_score_gemma":0.00002151,"domain_scores_codex":[0.9971521,0.0000154129,0.0004192853,0.001076911,0.0005476882,0.0007886123],"domain_scores_gemma":[0.9979364,0.0001397524,0.0002990454,0.00107504,0.0004597445,0.00009004714],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001348801,0.0003645513,0.01048485,0.0004344289,0.000120702,0.00002646324,0.01144166,0.00000430767,0.1196953,0.758559,0.005328249,0.09340557],"study_design_scores_gemma":[0.0005132143,0.0009348616,0.0004897563,0.0004038401,0.0000166353,0.00004983163,0.001124259,0.01741265,0.6089785,0.3175822,0.05167248,0.0008216876],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8803282,0.000428691,0.07958185,0.01519043,0.0003777623,0.0005317765,0.000004677262,0.01471737,0.008839179],"genre_scores_gemma":[0.9019129,0.0001247852,0.09706999,0.0006956832,0.00008440646,0.0000553576,0.000003310163,0.00002910349,0.0000245123],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4892832,"threshold_uncertainty_score":0.9998756,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01662893650988808,"score_gpt":0.2558164311824623,"score_spread":0.2391874946725742,"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."}}