{"id":"W4382469227","doi":"10.1609/aaai.v37i1.25088","title":"Denoising after Entropy-Based Debiasing a Robust Training Method for Dataset Bias with Noisy Labels","year":2023,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea","keywords":"Debiasing; Computer science; Artificial intelligence; Noise reduction; Machine learning; Generalization; Entropy (arrow of time); Sampling bias; Sample (material); Pattern recognition (psychology); Statistics; Mathematics; Sample size determination; Psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003649372,0.001132528,0.001368921,0.0008450348,0.0006657315,0.0009309696,0.001694718,0.001284362,0.001728542],"category_scores_gemma":[0.01251642,0.0006115504,0.001103775,0.0006129696,0.001410887,0.001796034,0.001774835,0.002392361,0.00076539],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006880954,"about_ca_system_score_gemma":0.001206987,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00225485,"about_ca_topic_score_gemma":0.003039905,"domain_scores_codex":[0.9985714,0.0002608527,0.0001288924,0.0005061273,0.0003982361,0.0001344745],"domain_scores_gemma":[0.9957299,0.001548184,0.0004812544,0.001127166,0.0009784555,0.0001351545],"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":[0.000733668,0.0002241098,0.008096849,0.0004744605,0.0002339438,0.0002897818,0.001139182,0.3354442,0.1006997,0.01236723,0.004729615,0.5355673],"study_design_scores_gemma":[0.00002270474,0.0001149677,0.001845654,0.00003828764,0.00004275093,0.0001160213,0.00005902155,0.9494956,0.04011278,0.005872396,0.002241669,0.00003827424],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03130788,0.0004054198,0.9655406,0.000201285,0.00006616743,0.00008015687,0.00008668529,0.001536642,0.0007751756],"genre_scores_gemma":[0.4124636,0.0003627594,0.5807418,0.0006065565,0.0001155222,0.0003259952,0.000827517,0.0006370106,0.003919231],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003649372,"threshold_uncertainty_score":0.01929992,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2141292935997887,"score_gpt":0.3536692546631078,"score_spread":0.1395399610633191,"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."}}