{"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/on-self-modulation-for-generative-adversarial","title":"On Self Modulation for Generative Adversarial Networks","arxiv_id":"1810.01365","date":"2018-10-02","proceeding":"ICLR 2019 5","authors":["Ting Chen","Mario Lucic","Neil Houlsby","Sylvain Gelly"],"abstract":"Training Generative Adversarial Networks (GANs) is notoriously challenging.\nWe propose and study an architectural modification, self-modulation, which\nimproves GAN performance across different data sets, architectures, losses,\nregularizers, and hyperparameter settings. Intuitively, self-modulation allows\nthe intermediate feature maps of a generator to change as a function of the\ninput noise vector. While reminiscent of other conditioning techniques, it\nrequires no labeled data. In a large-scale empirical study we observe a\nrelative decrease of $5\\%-35\\%$ in FID. Furthermore, all else being equal,\nadding this modification to the generator leads to improved performance in\n$124/144$ ($86\\%$) of the studied settings. Self-modulation is a simple\narchitectural change that requires no additional parameter tuning, which\nsuggests that it can be applied readily to any GAN.","url_abs":"http://arxiv.org/abs/1810.01365v2","url_pdf":"http://arxiv.org/pdf/1810.01365v2.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":"on-self-modulation-for-generative-adversarial","repo_url":"https://github.com/google/compare_gan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"on-self-modulation-for-generative-adversarial","repo_url":"https://github.com/kiban/models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1810.01365","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}