{"id":"W4382318343","doi":"10.1609/aaai.v37i12.26751","title":"Rethinking Label Refurbishment: Model Robustness under Label Noise","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":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"Robustness (evolution); Generalization; Computer science; Regularization (linguistics); Deep neural networks; Artificial neural network; Artificial intelligence; Noise (video); 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0114642,0.001923167,0.001734295,0.0009829496,0.001263598,0.002425197,0.004789218,0.003476207,0.001334882],"category_scores_gemma":[0.04120133,0.000914296,0.001676663,0.0007229663,0.003477373,0.005343598,0.004426373,0.006615682,0.0005593392],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001827816,"about_ca_system_score_gemma":0.001839559,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0077183,"about_ca_topic_score_gemma":0.00724839,"domain_scores_codex":[0.9958363,0.00190874,0.0001725325,0.001171941,0.0006197559,0.0002907886],"domain_scores_gemma":[0.9795154,0.01206359,0.001763503,0.004929229,0.001281991,0.0004462257],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009613781,0.0003222467,0.01167381,0.0002226893,0.0002793131,0.000283989,0.0008318559,0.8043935,0.00981494,0.02759636,0.004627426,0.1389925],"study_design_scores_gemma":[0.00003667264,0.0000494024,0.0003930712,0.00001978144,0.00001783422,0.00004088383,0.00002433354,0.9783589,0.001922425,0.01861119,0.0005051359,0.00002032232],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1163751,0.0009998672,0.8768237,0.001749106,0.0001456662,0.0001272205,0.0002129654,0.002162491,0.001404027],"genre_scores_gemma":[0.8227336,0.0003938711,0.1715006,0.001259876,0.0001905696,0.0001945044,0.0006428421,0.0005747207,0.002509451],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0114642,"threshold_uncertainty_score":0.06062919,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1653035614116124,"score_gpt":0.3342900065389262,"score_spread":0.1689864451273138,"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."}}