{"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-gans-with-optimism","title":"Training GANs with Optimism","arxiv_id":"1711.00141","date":"2017-10-31","proceeding":"ICLR 2018 1","authors":["Constantinos Daskalakis","Andrew Ilyas","Vasilis Syrgkanis","Haoyang Zeng"],"abstract":"We address the issue of limit cycling behavior in training Generative\nAdversarial Networks and propose the use of Optimistic Mirror Decent (OMD) for\ntraining Wasserstein GANs. Recent theoretical results have shown that\noptimistic mirror decent (OMD) can enjoy faster regret rates in the context of\nzero-sum games. WGANs is exactly a context of solving a zero-sum game with\nsimultaneous no-regret dynamics. Moreover, we show that optimistic mirror\ndecent addresses the limit cycling problem in training WGANs. We formally show\nthat in the case of bi-linear zero-sum games the last iterate of OMD dynamics\nconverges to an equilibrium, in contrast to GD dynamics which are bound to\ncycle. We also portray the huge qualitative difference between GD and OMD\ndynamics with toy examples, even when GD is modified with many adaptations\nproposed in the recent literature, such as gradient penalty or momentum. We\napply OMD WGAN training to a bioinformatics problem of generating DNA\nsequences. We observe that models trained with OMD achieve consistently smaller\nKL divergence with respect to the true underlying distribution, than models\ntrained with GD variants. Finally, we introduce a new algorithm, Optimistic\nAdam, which is an optimistic variant of Adam. We apply it to WGAN training on\nCIFAR10 and observe improved performance in terms of inception score as\ncompared to Adam.","url_abs":"http://arxiv.org/abs/1711.00141v2","url_pdf":"http://arxiv.org/pdf/1711.00141v2.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-gans-with-optimism","repo_url":"https://github.com/vsyrgkanis/optimistic_GAN_training","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"wgan","method_name":"WGAN"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.00141","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1711.00141"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vsyrgkanis/optimistic_GAN_training","reach":null}],"summary":{"ran_fixture":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"9be1c8d51fbe60bb","entry":"tile_images","repo":"vsyrgkanis/optimistic_GAN_training","repo_kind":"official","path":"script/cifar10.py","file_url":"https://github.com/vsyrgkanis/optimistic_GAN_training/blob/HEAD/script/cifar10.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"9be1c8d51fbe60bb"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}