{"id":"W4362454698","doi":"10.1016/j.neucom.2023.03.059","title":"LIAAD: Lightweight attentive angular distillation for large-scale age-invariant face recognition","year":2023,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Face recognition and analysis","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"Arkansas Biosciences Institute; National Science Foundation","keywords":"Computer science; Robustness (evolution); Facial recognition system; Artificial intelligence; Invariant (physics); Distillation; Face (sociological concept); Scale (ratio); Machine learning; Pattern recognition (psychology); Mathematics","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":[0.0004106811,0.0001745953,0.0002164781,0.0002621392,0.0004167676,0.0002578694,0.0003308627,0.00007191944,0.00001338121],"category_scores_gemma":[0.0001002362,0.0001739984,0.0002020734,0.001152053,0.00001384503,0.0003110526,0.0002078802,0.0001290766,0.000384821],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002760679,"about_ca_system_score_gemma":0.00002116773,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004338007,"about_ca_topic_score_gemma":0.000008568935,"domain_scores_codex":[0.9982662,0.0001014279,0.0003135095,0.0006083691,0.0002616978,0.0004488421],"domain_scores_gemma":[0.999073,0.0002420388,0.0001587281,0.0002699721,0.0001504041,0.0001058255],"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":[0.00005320257,0.0006820987,0.002321903,0.0004707127,0.0003307264,0.0004478412,0.01372829,0.009763218,0.04250954,0.007348061,0.01846452,0.9038799],"study_design_scores_gemma":[0.0005707916,0.00005875208,0.003312109,0.00006361646,0.00003242278,0.000009331176,0.0001536921,0.9788698,0.004596445,0.002118697,0.009921636,0.0002927289],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09349251,0.00001490884,0.9028534,0.001403438,0.000446603,0.0003784436,0.00003028909,0.0007643709,0.0006159878],"genre_scores_gemma":[0.9553362,0.00002366803,0.04210082,0.0008773707,0.0004240433,0.00005524978,0.0003149797,0.00003943707,0.0008281922],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9691066,"threshold_uncertainty_score":0.7095448,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03022797321782364,"score_gpt":0.2641729207251929,"score_spread":0.2339449475073693,"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."}}