{"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/k-beam-minimax-efficient-optimization-for","title":"K-Beam Minimax: Efficient Optimization for Deep Adversarial Learning","arxiv_id":"1805.11640","date":"2018-05-29","proceeding":"ICML 2018 7","authors":["Jihun Hamm","Yung-Kyun Noh"],"abstract":"Minimax optimization plays a key role in adversarial training of machine\nlearning algorithms, such as learning generative models, domain adaptation,\nprivacy preservation, and robust learning. In this paper, we demonstrate the\nfailure of alternating gradient descent in minimax optimization problems due to\nthe discontinuity of solutions of the inner maximization. To address this, we\npropose a new epsilon-subgradient descent algorithm that addresses this problem\nby simultaneously tracking K candidate solutions. Practically, the algorithm\ncan find solutions that previous saddle-point algorithms cannot find, with only\na sublinear increase of complexity in K. We analyze the conditions under which\nthe algorithm converges to the true solution in detail. A significant\nimprovement in stability and convergence speed of the algorithm is observed in\nsimple representative problems, GAN training, and domain-adaptation problems.","url_abs":"http://arxiv.org/abs/1805.11640v2","url_pdf":"http://arxiv.org/pdf/1805.11640v2.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":"k-beam-minimax-efficient-optimization-for","repo_url":"https://github.com/jihunhamm/k-beam-minimax","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}