{"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/the-unusual-effectiveness-of-averaging-in-gan","title":"The Unusual Effectiveness of Averaging in GAN Training","arxiv_id":"1806.04498","date":"2018-06-12","proceeding":"ICLR 2019 5","authors":["Yasin Yazici","Chuan-Sheng Foo","Stefan Winkler","Kim-Hui Yap","Georgios Piliouras","Vijay Chandrasekhar"],"abstract":"We examine two different techniques for parameter averaging in GAN training.\nMoving Average (MA) computes the time-average of parameters, whereas\nExponential Moving Average (EMA) computes an exponentially discounted sum.\nWhilst MA is known to lead to convergence in bilinear settings, we provide the\n-- to our knowledge -- first theoretical arguments in support of EMA. We show\nthat EMA converges to limit cycles around the equilibrium with vanishing\namplitude as the discount parameter approaches one for simple bilinear games\nand also enhances the stability of general GAN training. We establish\nexperimentally that both techniques are strikingly effective in the\nnon-convex-concave GAN setting as well. Both improve inception and FID scores\non different architectures and for different GAN objectives. We provide\ncomprehensive experimental results across a range of datasets -- mixture of\nGaussians, CIFAR-10, STL-10, CelebA and ImageNet -- to demonstrate its\neffectiveness. We achieve state-of-the-art results on CIFAR-10 and produce\nclean CelebA face images.\\footnote{~The code is available at\n\\url{https://github.com/yasinyazici/EMA_GAN}}","url_abs":"http://arxiv.org/abs/1806.04498v2","url_pdf":"http://arxiv.org/pdf/1806.04498v2.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":"the-unusual-effectiveness-of-averaging-in-gan","repo_url":"https://github.com/yasinyazici/EMA_GAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.04498","atlas_url":"https://app.syntology.ai/?focus=1806.04498","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1806.04498"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/yasinyazici/EMA_GAN","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":3},"by_repo_kind":{"official":{"samples":3,"ran":0,"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":0,"samples":[{"code_sha256_prefix":"886c94110552d89a","entry":"image_to_np","repo":"yasinyazici/EMA_GAN","repo_kind":"official","path":"common/dataset.py","file_url":"https://github.com/yasinyazici/EMA_GAN/blob/HEAD/common/dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"886c94110552d89a"}},{"code_sha256_prefix":"4bb6318a05fd4d11","entry":"preprocess_image","repo":"yasinyazici/EMA_GAN","repo_kind":"official","path":"common/dataset.py","file_url":"https://github.com/yasinyazici/EMA_GAN/blob/HEAD/common/dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4bb6318a05fd4d11"}},{"code_sha256_prefix":"61285bf457777aae","entry":"preprocess_image_no_crop","repo":"yasinyazici/EMA_GAN","repo_kind":"official","path":"common/dataset.py","file_url":"https://github.com/yasinyazici/EMA_GAN/blob/HEAD/common/dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"61285bf457777aae"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}