{"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/generative-adversarial-nets-for-robust","title":"Generative Adversarial Nets for Robust Scatter Estimation: A Proper Scoring Rule Perspective","arxiv_id":"1903.01944","date":"2019-03-05","proceeding":null,"authors":["Chao Gao","Yuan YAO","Weizhi Zhu"],"abstract":"Robust scatter estimation is a fundamental task in statistics. The recent\ndiscovery on the connection between robust estimation and generative\nadversarial nets (GANs) by Gao et al. (2018) suggests that it is possible to\ncompute depth-like robust estimators using similar techniques that optimize\nGANs. In this paper, we introduce a general learning via classification\nframework based on the notion of proper scoring rules. This framework allows us\nto understand both matrix depth function and various GANs through the lens of\nvariational approximations of $f$-divergences induced by proper scoring rules.\nWe then propose a new class of robust scatter estimators in this framework by\ncarefully constructing discriminators with appropriate neural network\nstructures. These estimators are proved to achieve the minimax rate of scatter\nestimation under Huber's contamination model. Our numerical results demonstrate\nits good performance under various settings against competitors in the\nliterature.","url_abs":"http://arxiv.org/abs/1903.01944v1","url_pdf":"http://arxiv.org/pdf/1903.01944v1.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":"generative-adversarial-nets-for-robust","repo_url":"https://github.com/zhuwzh/Robust-GAN-Scatter","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"scoring-rule","task_name":"scoring rule"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.01944","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}