{"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/composite-functional-gradient-learning-of","title":"Composite Functional Gradient Learning of Generative Adversarial Models","arxiv_id":"1801.06309","date":"2018-01-19","proceeding":"ICML 2018 7","authors":["Rie Johnson","Tong Zhang"],"abstract":"This paper first presents a theory for generative adversarial methods that\ndoes not rely on the traditional minimax formulation. It shows that with a\nstrong discriminator, a good generator can be learned so that the KL divergence\nbetween the distributions of real data and generated data improves after each\nfunctional gradient step until it converges to zero. Based on the theory, we\npropose a new stable generative adversarial method. A theoretical insight into\nthe original GAN from this new viewpoint is also provided. The experiments on\nimage generation show the effectiveness of our new method.","url_abs":"http://arxiv.org/abs/1801.06309v2","url_pdf":"http://arxiv.org/pdf/1801.06309v2.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":"composite-functional-gradient-learning-of","repo_url":"https://github.com/riejohnson/cfg-gan-pt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}