{"id":"W2970987521","doi":"10.1007/978-3-030-26250-1_23","title":"Improving ML Safety with Partial Specifications","year":2019,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Workflow; Software engineering; Key (lock); Component (thermodynamics); Safety assurance; Software; Partial evaluation; Programming language; Reliability engineering; Operating system; Database; 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":[0.004788127,0.00146034,0.001231731,0.0009434241,0.0006746562,0.002614164,0.002154695,0.001791667,0.008451843],"category_scores_gemma":[0.02682595,0.001115091,0.001759461,0.0006480427,0.003468386,0.005022935,0.005919332,0.005394058,0.002552529],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008366418,"about_ca_system_score_gemma":0.001717582,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000541391,"about_ca_topic_score_gemma":0.000788825,"domain_scores_codex":[0.9918495,0.002741053,0.0005128754,0.0009092209,0.003323137,0.0006641211],"domain_scores_gemma":[0.9750591,0.01501928,0.0008716194,0.00717147,0.001585277,0.0002932415],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008191855,0.000187695,0.00159013,0.000546834,0.0001688398,0.0003632948,0.0003567699,0.2672623,0.0231271,0.3726676,0.01031298,0.3225972],"study_design_scores_gemma":[0.00005625892,0.000201276,0.0001409446,0.00009363158,0.00006182217,0.0001924361,0.00004856829,0.5186642,0.02273456,0.4517358,0.006030155,0.00004028904],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01334218,0.0003269618,0.9756156,0.0005475804,0.0001053932,0.00004191677,0.000123632,0.003645315,0.006251452],"genre_scores_gemma":[0.691106,0.0005702673,0.2932787,0.0009691049,0.000382863,0.0001656017,0.0008442718,0.002431439,0.01025183],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008451843,"threshold_uncertainty_score":0.02827424,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01840434861093901,"score_gpt":0.2362812111301523,"score_spread":0.2178768625192133,"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."}}