{"id":"W4287905575","doi":"10.48550/arxiv.2001.03071","title":"Investigating the Impact of Inclusion in Face Recognition Training Data\\n on Individual Face Identification","year":2020,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Face recognition and analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Facial recognition system; Computer science; Artificial intelligence; Identification (biology); Convolutional neural network; Three-dimensional face recognition; Face Recognition Grand Challenge; Embedding; Pattern recognition (psychology); Machine learning; Face detection","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":[0.002327489,0.0005518934,0.0006997427,0.0007271338,0.0007057212,0.0003440993,0.004989031,0.0003788039,0.00008369358],"category_scores_gemma":[0.0008006091,0.0005276984,0.0004457758,0.003570786,0.0003549223,0.001284445,0.0127627,0.001449862,0.0001551676],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003764111,"about_ca_system_score_gemma":0.0007020489,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000908713,"about_ca_topic_score_gemma":0.0001476096,"domain_scores_codex":[0.9948253,0.001193442,0.0009098966,0.002114992,0.0004605768,0.0004957899],"domain_scores_gemma":[0.9953721,0.000611125,0.001550262,0.001901372,0.0002749853,0.000290152],"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":[0.00008151263,0.0004556169,0.00263545,0.0001943969,0.0006426736,0.00007113516,0.04102528,0.825852,0.003226402,0.001573225,0.0001823403,0.1240599],"study_design_scores_gemma":[0.0007542468,0.0001499582,0.009410823,0.0007120661,0.0002405673,0.000003261092,0.004918262,0.958869,0.0009088563,0.02347888,0.00001001334,0.0005441081],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7847659,0.00004026032,0.2124005,0.0008606759,0.0002041096,0.0006380435,0.000536811,0.00007709893,0.0004766212],"genre_scores_gemma":[0.9978842,0.0003524103,0.0004755133,0.0001611885,0.00006497923,0.000001617152,0.0009620937,0.00002500849,0.00007301602],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2131183,"threshold_uncertainty_score":0.9997175,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3142147760585666,"score_gpt":0.2694955620556808,"score_spread":0.04471921400288581,"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."}}