{"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/an-online-learning-approach-to-generative","title":"An Online Learning Approach to Generative Adversarial Networks","arxiv_id":"1706.03269","date":"2017-06-10","proceeding":"ICLR 2018 1","authors":["Paulina Grnarova","Kfir. Y. Levy","Aurelien Lucchi","Thomas Hofmann","Andreas Krause"],"abstract":"We consider the problem of training generative models with a Generative\nAdversarial Network (GAN). Although GANs can accurately model complex\ndistributions, they are known to be difficult to train due to instabilities\ncaused by a difficult minimax optimization problem. In this paper, we view the\nproblem of training GANs as finding a mixed strategy in a zero-sum game.\nBuilding on ideas from online learning we propose a novel training method named\nChekhov GAN 1 . On the theory side, we show that our method provably converges\nto an equilibrium for semi-shallow GAN architectures, i.e. architectures where\nthe discriminator is a one layer network and the generator is arbitrary. On the\npractical side, we develop an efficient heuristic guided by our theoretical\nresults, which we apply to commonly used deep GAN architectures. On several\nreal world tasks our approach exhibits improved stability and performance\ncompared to standard GAN training.","url_abs":"http://arxiv.org/abs/1706.03269v1","url_pdf":"http://arxiv.org/pdf/1706.03269v1.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":"an-online-learning-approach-to-generative","repo_url":"https://github.com/vaishn99/modified-GAN-training","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.03269","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}