{"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/a-minimax-surrogate-loss-approach-to","title":"A Minimax Surrogate Loss Approach to Conditional Difference Estimation","arxiv_id":"1803.03769","date":"2018-03-10","proceeding":null,"authors":["Siong Thye Goh","Cynthia Rudin"],"abstract":"We present a new machine learning approach to estimate personalized treatment\neffects in the classical potential outcomes framework with binary outcomes. To\novercome the problem that both treatment and control outcomes for the same unit\nare required for supervised learning, we propose surrogate loss functions that\nincorporate both treatment and control data. The new surrogates yield tighter\nbounds than the sum of losses for treatment and control groups. A specific\nchoice of loss function, namely a type of hinge loss, yields a minimax support\nvector machine formulation. The resulting optimization problem requires the\nsolution to only a single convex optimization problem, incorporating both\ntreatment and control units, and it enables the kernel trick to be used to\nhandle nonlinear (also non-parametric) estimation. Statistical learning bounds\nare also presented for the framework, and experimental results.","url_abs":"http://arxiv.org/abs/1803.03769v2","url_pdf":"http://arxiv.org/pdf/1803.03769v2.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":"a-minimax-surrogate-loss-approach-to","repo_url":"https://github.com/shangtai/githubcausalsvm","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}