{"id":"W4379767400","doi":"10.1007/978-3-031-34048-2_13","title":"On Fairness of Medical Image Classification with Multiple Sensitive Attributes via Learning Orthogonal Representations","year":2023,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":13,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of British Columbia","funders":"","keywords":"Orthogonality; Computer science; Representation (politics); Orthogonal complement; Space (punctuation); Artificial intelligence; Complement (music); Field (mathematics); Code (set theory); Independence (probability theory); Feature (linguistics); Feature learning; Machine learning; Feature vector; Rank (graph theory); Pattern recognition (psychology); Image (mathematics); Contextual image classification; Focus (optics); 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":[0.03167967,0.00123481,0.004042364,0.001823409,0.001792067,0.004148077,0.004850056,0.002976096,0.004161088],"category_scores_gemma":[0.08767369,0.0009726463,0.001594926,0.002364163,0.007616083,0.009962525,0.009050888,0.005265371,0.0005024296],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004612422,"about_ca_system_score_gemma":0.003543212,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003642346,"about_ca_topic_score_gemma":0.001997019,"domain_scores_codex":[0.978626,0.01276119,0.0008250599,0.002876052,0.003596963,0.001314666],"domain_scores_gemma":[0.8957038,0.08393604,0.00353662,0.01019922,0.004656347,0.00196802],"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":[0.001275514,0.0002107742,0.003401852,0.000210291,0.0001604434,0.0001327879,0.0004183934,0.193096,0.001861603,0.6857989,0.005335893,0.1080976],"study_design_scores_gemma":[0.0000302114,0.0000624784,0.0003545315,0.00002684962,0.00002284185,0.00004476531,0.00003799643,0.540865,0.0006466876,0.4572911,0.0005987576,0.00001888981],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02920303,0.0006104417,0.9650702,0.001595104,0.0001408167,0.00008800715,0.0001175267,0.0001423894,0.003032378],"genre_scores_gemma":[0.7210841,0.0008741344,0.2651554,0.0009049565,0.0009053833,0.0002695612,0.0003679621,0.00030333,0.01013527],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.03167967,"threshold_uncertainty_score":0.1675401,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1098935778929578,"score_gpt":0.3893378997809668,"score_spread":0.2794443218880089,"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."}}