{"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/training-generative-networks-using-random","title":"Training generative networks using random discriminators","arxiv_id":"1904.09775","date":"2019-04-22","proceeding":null,"authors":["Babak Barazandeh","Meisam Razaviyayn","Maziar Sanjabi"],"abstract":"In recent years, Generative Adversarial Networks (GANs) have drawn a lot of\nattentions for learning the underlying distribution of data in various\napplications. Despite their wide applicability, training GANs is notoriously\ndifficult. This difficulty is due to the min-max nature of the resulting\noptimization problem and the lack of proper tools of solving general\n(non-convex, non-concave) min-max optimization problems. In this paper, we try\nto alleviate this problem by proposing a new generative network that relies on\nthe use of random discriminators instead of adversarial design. This design\nhelps us to avoid the min-max formulation and leads to an optimization problem\nthat is stable and could be solved efficiently. The performance of the proposed\nmethod is evaluated using handwritten digits (MNIST) and Fashion products\n(Fashion-MNIST) data sets. While the resulting images are not as sharp as\nadversarial training, the use of random discriminator leads to a much faster\nalgorithm as compared to the adversarial counterpart. This observation, at the\nminimum, illustrates the potential of the random discriminator approach for\nwarm-start in training GANs.","url_abs":"http://arxiv.org/abs/1904.09775v1","url_pdf":"http://arxiv.org/pdf/1904.09775v1.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":"training-generative-networks-using-random","repo_url":"https://github.com/babakbarazandeh/GN-RD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"training-generative-networks-using-random","repo_url":"https://github.com/optimization-for-data-driven-science/GN-RD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","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}