{"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/fast-threshold-tests-for-detecting","title":"Fast Threshold Tests for Detecting Discrimination","arxiv_id":"1702.08536","date":"2017-02-27","proceeding":null,"authors":["Emma Pierson","Sam Corbett-Davies","Sharad Goel"],"abstract":"Threshold tests have recently been proposed as a useful method for detecting\nbias in lending, hiring, and policing decisions. For example, in the case of\ncredit extensions, these tests aim to estimate the bar for granting loans to\nwhite and minority applicants, with a higher inferred threshold for minorities\nindicative of discrimination. This technique, however, requires fitting a\ncomplex Bayesian latent variable model for which inference is often\ncomputationally challenging. Here we develop a method for fitting threshold\ntests that is two orders of magnitude faster than the existing approach,\nreducing computation from hours to minutes. To achieve these performance gains,\nwe introduce and analyze a flexible family of probability distributions on the\ninterval [0, 1] -- which we call discriminant distributions -- that is\ncomputationally efficient to work with. We demonstrate our technique by\nanalyzing 2.7 million police stops of pedestrians in New York City.","url_abs":"http://arxiv.org/abs/1702.08536v3","url_pdf":"http://arxiv.org/pdf/1702.08536v3.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":"fast-threshold-tests-for-detecting","repo_url":"https://github.com/5harad/fasttt","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1702.08536","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}