{"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/achieving-fairness-through-adversarial","title":"Achieving Fairness through Adversarial Learning: an Application to Recidivism Prediction","arxiv_id":"1807.00199","date":"2018-06-30","proceeding":null,"authors":["Christina Wadsworth","Francesca Vera","Chris Piech"],"abstract":"Recidivism prediction scores are used across the USA to determine sentencing\nand supervision for hundreds of thousands of inmates. One such generator of\nrecidivism prediction scores is Northpointe's Correctional Offender Management\nProfiling for Alternative Sanctions (COMPAS) score, used in states like\nCalifornia and Florida, which past research has shown to be biased against\nblack inmates according to certain measures of fairness. To counteract this\nracial bias, we present an adversarially-trained neural network that predicts\nrecidivism and is trained to remove racial bias. When comparing the results of\nour model to COMPAS, we gain predictive accuracy and get closer to achieving\ntwo out of three measures of fairness: parity and equality of odds. Our model\ncan be generalized to any prediction and demographic. This piece of research\ncontributes an example of scientific replication and simplification in a\nhigh-stakes real-world application like recidivism prediction.","url_abs":"http://arxiv.org/abs/1807.00199v1","url_pdf":"http://arxiv.org/pdf/1807.00199v1.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":"achieving-fairness-through-adversarial","repo_url":"https://github.com/dns43/fairness","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"achieving-fairness-through-adversarial","repo_url":"https://github.com/istaustria-cvml/flea","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":null}],"tasks":[{"task_slug":"fairness","task_name":"Fairness"},{"task_slug":"management","task_name":"Management"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.00199","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}