{"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/lipizzaner-a-system-that-scales-robust","title":"Lipizzaner: A System That Scales Robust Generative Adversarial Network Training","arxiv_id":"1811.12843","date":"2018-11-30","proceeding":null,"authors":["Tom Schmiedlechner","Ignavier Ng Zhi Yong","Abdullah Al-Dujaili","Erik Hemberg","Una-May O'Reilly"],"abstract":"GANs are difficult to train due to convergence pathologies such as mode and\ndiscriminator collapse. We introduce Lipizzaner, an open source software system\nthat allows machine learning engineers to train GANs in a distributed and\nrobust way. Lipizzaner distributes a competitive coevolutionary algorithm\nwhich, by virtue of dual, adapting, generator and discriminator populations, is\nrobust to collapses. The algorithm is well suited to efficient distribution\nbecause it uses a spatial grid abstraction. Training is local to each cell and\nstrong intermediate training results are exchanged among overlapping\nneighborhoods allowing high performing solutions to propagate and improve with\nmore rounds of training. Experiments on common image datasets overcome critical\ncollapses. Communication overhead scales linearly when increasing the number of\ncompute instances and we observe that increasing scale leads to improved model\nperformance.","url_abs":"http://arxiv.org/abs/1811.12843v1","url_pdf":"http://arxiv.org/pdf/1811.12843v1.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":"lipizzaner-a-system-that-scales-robust","repo_url":"https://github.com/ALFA-group/lipizzaner-gan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}