{"id":"W3205783011","doi":"","title":"Tighter Risk Certificates for Neural Networks","year":2020,"lang":"en","type":"article","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Army Research Laboratory; Army Research Office; Engineering and Physical Sciences Research Council; University College London; DeepMind; Alberta Machine Intelligence Institute; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Artificial neural network; Artificial intelligence; Machine learning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001170682,0.0001463685,0.0001494266,0.00005702275,0.0004027783,0.0004216795,0.0008558054,0.00007616545,0.00008331184],"category_scores_gemma":[0.0004600364,0.000144313,0.0001327144,0.0003332183,0.0000652011,0.000396743,0.000209645,0.0002001339,0.00006866546],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001895249,"about_ca_system_score_gemma":0.00002833765,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005008936,"about_ca_topic_score_gemma":0.00006066424,"domain_scores_codex":[0.9976239,0.001192327,0.0002775989,0.0004788314,0.0001685555,0.0002588061],"domain_scores_gemma":[0.9974396,0.0007999776,0.000209372,0.0006575231,0.0007110581,0.0001824707],"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":[0.00003719364,0.000517427,0.002645932,0.0000788745,0.0001036913,0.000005994677,0.01292363,0.001032607,0.004855436,0.1554706,0.02624371,0.7960849],"study_design_scores_gemma":[0.0004335636,0.00000100894,0.001227443,0.0000502133,0.00001538439,0.000003546768,0.00002512336,0.9345839,0.02824469,0.002417359,0.03277598,0.0002218016],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01848435,0.0001676523,0.9469517,0.02896151,0.0001367498,0.0002504008,0.00001254643,0.00033044,0.004704593],"genre_scores_gemma":[0.9525061,0.00008065446,0.04538891,0.0008794396,0.00005931855,0.00004889057,0.00007686495,0.00001794211,0.0009419211],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9340217,"threshold_uncertainty_score":0.5884912,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02164719154634548,"score_gpt":0.2178465939448264,"score_spread":0.1961994023984809,"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."}}