{"id":"W4382317875","doi":"10.1609/aaai.v37i12.26748","title":"Revisiting the Importance of Amplifying Bias for Debiasing","year":2023,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Ministry of Science and ICT, South Korea; Korea Advanced Institute of Science and Technology; Institute for Information and Communications Technology Promotion; Electronics and Telecommunications Research Institute","keywords":"Debiasing; Selection bias; Computer science; Artificial intelligence; Classifier (UML); Sampling bias; Non-response bias; Training set; Sample (material); Statistics; Machine learning; Sample size determination; Psychology; Mathematics; Social 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002119106,0.0002039863,0.0003058954,0.0001405839,0.0004539175,0.0002016098,0.002637386,0.00007311531,0.00001204271],"category_scores_gemma":[0.004076228,0.0001350547,0.0001851098,0.001425009,0.0002662233,0.0003717313,0.0006825245,0.0003345732,0.00001767872],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003644094,"about_ca_system_score_gemma":0.00009549936,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002116425,"about_ca_topic_score_gemma":0.000003318728,"domain_scores_codex":[0.9978384,0.00003418226,0.0007385224,0.0004609717,0.0005135876,0.0004143549],"domain_scores_gemma":[0.9970863,0.0008737678,0.0009072527,0.000410956,0.0006738498,0.0000478875],"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":[0.00003183284,0.00001331587,0.003000482,0.0001116352,0.00001630903,3.013363e-7,0.001719395,0.002247259,0.01337915,0.8756062,0.00005654685,0.1038176],"study_design_scores_gemma":[0.00004340738,0.00008393869,0.001143686,0.0006713714,0.00002320061,0.000004460606,0.001898024,0.5818698,0.1771234,0.2366902,0.0001823163,0.0002662194],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4348817,0.00006241108,0.5375334,0.01916605,0.001104963,0.001360526,0.000008579648,0.0003909766,0.005491458],"genre_scores_gemma":[0.9885328,0.00001995071,0.01095864,0.0001594826,0.0001791292,0.00003445591,5.330147e-7,0.00001852046,0.00009644755],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.638916,"threshold_uncertainty_score":0.5507373,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1840778661745962,"score_gpt":0.3545708136863175,"score_spread":0.1704929475117214,"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."}}