{"id":"W4388213050","doi":"10.32614/rj-2023-034","title":"Fairness Audits and Debiasing Using \\pkg{mlr3fairness}","year":2023,"lang":"en","type":"article","venue":"The R Journal","topic":"Privacy, Security, and Data Protection","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Debiasing; Computer science; Audit; Variety (cybernetics); Machine learning; Set (abstract data type); Artificial intelligence; Data science; Accounting","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":[0.04658337,0.0021942,0.001929059,0.004162811,0.002473616,0.007380961,0.003801678,0.003126948,0.02034032],"category_scores_gemma":[0.2139515,0.001207263,0.002039345,0.002466068,0.003290972,0.0081636,0.008081229,0.005115394,0.008558193],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002709692,"about_ca_system_score_gemma":0.005405941,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00612216,"about_ca_topic_score_gemma":0.005324927,"domain_scores_codex":[0.944523,0.02716245,0.003862925,0.004840407,0.01715456,0.002456704],"domain_scores_gemma":[0.830507,0.08877493,0.01266075,0.05060498,0.01503988,0.0024125],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002848187,0.0006085589,0.04055785,0.001393426,0.0005139569,0.0006871488,0.003281687,0.07665604,0.007691006,0.1354926,0.2453029,0.4849668],"study_design_scores_gemma":[0.0003622929,0.0004064753,0.007920044,0.0007076695,0.0001177915,0.0006071322,0.0005012914,0.5930706,0.02986696,0.2566974,0.1092316,0.0005108517],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03751796,0.0008451157,0.6580636,0.004502446,0.0008542385,0.0007920342,0.006677136,0.2713757,0.01937173],"genre_scores_gemma":[0.4914238,0.0005340072,0.4324903,0.003215955,0.0004267094,0.001655929,0.007821976,0.04910758,0.01332368],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04658337,"threshold_uncertainty_score":0.2463593,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07956026539987977,"score_gpt":0.3494817351495638,"score_spread":0.2699214697496841,"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."}}