{"id":"W2143340430","doi":"10.1016/j.patrec.2004.09.014","title":"Regularization studies of linear discriminant analysis in small sample size scenarios with application to face recognition","year":2004,"lang":"en","type":"article","venue":"Pattern Recognition Letters","topic":"Face and Expression Recognition","field":"Computer Science","cited_by":306,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto; Bell (Canada)","funders":"National Institute of Standards and Technology","keywords":"Linear discriminant analysis; Pattern recognition (psychology); Artificial intelligence; Computer science; Regularization (linguistics); Facial recognition system; Curse of dimensionality; Sample size determination; Face (sociological concept); Eigenface; Principal component analysis; Mathematics; Machine learning; Statistics","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":[],"consensus_categories":[],"category_scores_codex":[0.0003158135,0.0002426275,0.0003858184,0.0006235949,0.0001126259,0.00005699629,0.0003003704,0.00008172162,0.00001541016],"category_scores_gemma":[0.000214392,0.0002167604,0.0001103889,0.001793781,0.00005578646,0.0005005716,0.00009694825,0.0001474295,0.00008668243],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001341728,"about_ca_system_score_gemma":0.00003019619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006247888,"about_ca_topic_score_gemma":0.001074018,"domain_scores_codex":[0.9980577,0.0001235568,0.0005447876,0.0006537552,0.0003080847,0.0003120812],"domain_scores_gemma":[0.9986181,0.0002806352,0.0003059362,0.0004109688,0.0002835834,0.00010073],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002587165,0.0009035758,0.01953046,0.000443523,0.0007220193,0.0000387528,0.01503692,0.05645462,0.1244422,0.00002709863,0.0001372132,0.7820049],"study_design_scores_gemma":[0.01419147,0.00193801,0.1890925,0.005657039,0.002272137,0.00007287593,0.008678962,0.07170805,0.6690942,0.03196691,0.000248538,0.005079229],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4518086,0.000009339423,0.5447928,0.002829812,0.00005112456,0.0004004807,0.00004214312,0.00005752684,0.000008183299],"genre_scores_gemma":[0.9038672,0.00004959158,0.09257568,0.002833699,0.00004768712,0.0002830326,0.0003182618,0.00001993734,0.000004884732],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7769257,"threshold_uncertainty_score":0.8839232,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0353282354177338,"score_gpt":0.2647489232530105,"score_spread":0.2294206878352767,"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."}}