{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001054659,0.0002368463,0.0002377114,0.0002132854,0.0003698989,0.0003895798,0.002569058,0.0001257789,0.0000177412],"category_scores_gemma":[0.0004666345,0.0001850128,0.00006957557,0.001426387,0.0001816247,0.0006118238,0.0006861153,0.0004806089,0.00019832],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005795565,"about_ca_system_score_gemma":0.0001244134,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003099617,"about_ca_topic_score_gemma":0.000008871837,"domain_scores_codex":[0.9976751,0.00002993313,0.0005239859,0.0006568528,0.0006915856,0.0004225479],"domain_scores_gemma":[0.9983022,0.0001109706,0.0004075372,0.0005483438,0.0005338556,0.00009713008],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002493557,0.00008695798,0.00008147529,0.00004336516,0.00001065067,3.962991e-7,0.001325221,0.006330182,0.02907811,0.9057194,0.0004578898,0.05684143],"study_design_scores_gemma":[0.00003212233,0.00004963099,0.0001274541,0.0001330439,0.000007504163,0.000001582533,0.0001566178,0.7311258,0.04542426,0.2227454,0.00002887474,0.0001677218],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5383455,0.0000452152,0.3793823,0.05169322,0.001251165,0.0009935828,0.00002657349,0.001563299,0.02669922],"genre_scores_gemma":[0.9887146,0.0000612398,0.009533771,0.0003397186,0.00005252416,0.00003301692,0.00000435122,0.00001861,0.001242151],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7247956,"threshold_uncertainty_score":0.7544601,"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."}}