{"id":"W4383604771","doi":"10.48550/arxiv.2307.02744","title":"Active Learning with Contrastive Pre-training for Facial Expression Recognition","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Mitacs","keywords":"Computer science; Context (archaeology); Artificial intelligence; Set (abstract data type); Machine learning; Active learning (machine learning); Baseline (sea); Code (set theory); Labelling; Labeled data; Deep learning; Test set; Training set; Source code; Natural language processing; Psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002967468,0.001395148,0.0010493,0.001041493,0.0004240703,0.001065447,0.00297334,0.001626668,0.003239801],"category_scores_gemma":[0.00668329,0.0006452177,0.001202572,0.0007385489,0.001138582,0.001947568,0.001441979,0.003329068,0.002173613],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001030668,"about_ca_system_score_gemma":0.0007425871,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002970807,"about_ca_topic_score_gemma":0.005637151,"domain_scores_codex":[0.9986545,0.00048966,0.00005346207,0.0003345244,0.0003185346,0.0001493139],"domain_scores_gemma":[0.9974725,0.001514687,0.0001346808,0.000455296,0.0003436408,0.00007917158],"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":[0.0006132658,0.0006612072,0.002011142,0.0002277543,0.0001669133,0.00009217275,0.0001856097,0.1852384,0.02486293,0.007334097,0.01167706,0.7669294],"study_design_scores_gemma":[0.00002424398,0.0001133683,0.0004673084,0.00002733373,0.00001794071,0.00005675992,0.00002057579,0.9754014,0.01585402,0.00511819,0.002882266,0.00001665147],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02783837,0.001061505,0.9616883,0.0004819438,0.0001550375,0.0001390129,0.0002620759,0.004551463,0.003822325],"genre_scores_gemma":[0.4636654,0.000716717,0.5226139,0.00100538,0.0001798784,0.000480926,0.002157387,0.0005837216,0.008596713],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003239801,"threshold_uncertainty_score":0.0156936,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1141270970978609,"score_gpt":0.2219462995758826,"score_spread":0.1078192024780217,"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."}}