{"id":"W4321019977","doi":"10.1016/j.patcog.2023.109417","title":"Timid semi–supervised learning for face expression analysis","year":2023,"lang":"en","type":"article","venue":"Pattern Recognition","topic":"Face and Expression Recognition","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Ontario Ministry of Research and Innovation; Unitatea Executiva pentru Finantarea Invatamantului Superior, a Cercetarii, Dezvoltarii si Inovarii; Nvidia","keywords":"Computer science; Artificial intelligence; Machine learning; Face (sociological concept); Supervised learning; Domain (mathematical analysis); Expression (computer science); Semi-supervised learning; Action (physics); Labeled data; Pattern recognition (psychology); Mathematics","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0003683219,0.0001687688,0.0002171864,0.0005097621,0.0002788436,0.0001757621,0.0003163483,0.0001165495,0.0001733884],"category_scores_gemma":[0.0000895097,0.0001606272,0.0002281021,0.001225425,0.00001310641,0.0005778132,0.0001411321,0.0001512293,0.001310922],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002442423,"about_ca_system_score_gemma":0.0000149912,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001591277,"about_ca_topic_score_gemma":0.000006208026,"domain_scores_codex":[0.9984315,0.000124075,0.0002699164,0.0005297638,0.0002823205,0.0003624204],"domain_scores_gemma":[0.9991105,0.0002363204,0.0001236576,0.0002760911,0.0001470319,0.0001064293],"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":[0.0000223737,0.0000714241,0.004104953,0.00009154397,0.000131583,0.000009751522,0.00129036,0.002608406,0.04096459,0.000002572429,0.007484916,0.9432175],"study_design_scores_gemma":[0.0009783378,0.0001283196,0.004870726,0.0001801282,0.0001512032,0.000003048272,0.0004231069,0.8792221,0.109645,0.00171271,0.002174001,0.0005112898],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2470776,0.00001505651,0.7509061,0.0004518716,0.0002441666,0.0002710303,0.00002722791,0.0006500037,0.0003569401],"genre_scores_gemma":[0.9920873,0.00007906256,0.005182744,0.000372519,0.0001385683,0.0002998383,0.001343787,0.00002222398,0.0004739791],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9427062,"threshold_uncertainty_score":0.9994667,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04501966001723489,"score_gpt":0.2769863194701571,"score_spread":0.2319666594529222,"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."}}