{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/multiaccuracy-black-box-post-processing-for","title":"Multiaccuracy: Black-Box Post-Processing for Fairness in Classification","arxiv_id":"1805.12317","date":"2018-05-31","proceeding":null,"authors":["Michael P. Kim","Amirata Ghorbani","James Zou"],"abstract":"Prediction systems are successfully deployed in applications ranging from\ndisease diagnosis, to predicting credit worthiness, to image recognition. Even\nwhen the overall accuracy is high, these systems may exhibit systematic biases\nthat harm specific subpopulations; such biases may arise inadvertently due to\nunderrepresentation in the data used to train a machine-learning model, or as\nthe result of intentional malicious discrimination. We develop a rigorous\nframework of *multiaccuracy* auditing and post-processing to ensure accurate\npredictions across *identifiable subgroups*.\n  Our algorithm, MULTIACCURACY-BOOST, works in any setting where we have\nblack-box access to a predictor and a relatively small set of labeled data for\nauditing; importantly, this black-box framework allows for improved fairness\nand accountability of predictions, even when the predictor is minimally\ntransparent. We prove that MULTIACCURACY-BOOST converges efficiently and show\nthat if the initial model is accurate on an identifiable subgroup, then the\npost-processed model will be also. We experimentally demonstrate the\neffectiveness of the approach to improve the accuracy among minority subgroups\nin diverse applications (image classification, finance, population health).\nInterestingly, MULTIACCURACY-BOOST can improve subpopulation accuracy (e.g. for\n\"black women\") even when the sensitive features (e.g. \"race\", \"gender\") are not\ngiven to the algorithm explicitly.","url_abs":"http://arxiv.org/abs/1805.12317v2","url_pdf":"http://arxiv.org/pdf/1805.12317v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"multiaccuracy-black-box-post-processing-for","repo_url":"https://github.com/amiratag/multiaccuracyboost","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"fairness","task_name":"Fairness"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.12317","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1805.12317"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/amiratag/multiaccuracyboost","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":3},"by_repo_kind":{"listed":{"samples":3,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"a202aee4c0d10ea5","entry":"get_session","repo":"amiratag/multiaccuracyboost","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/amiratag/multiaccuracyboost/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a202aee4c0d10ea5"}},{"code_sha256_prefix":"448d969a51966ef6","entry":"read_images","repo":"amiratag/multiaccuracyboost","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/amiratag/multiaccuracyboost/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"448d969a51966ef6"}},{"code_sha256_prefix":"fbdc5caf3f79baba","entry":"split_data","repo":"amiratag/multiaccuracyboost","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/amiratag/multiaccuracyboost/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fbdc5caf3f79baba"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}