{"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/multiple-source-domain-adaptation-with","title":"Multiple Source Domain Adaptation with Adversarial Training of Neural Networks","arxiv_id":"1705.09684","date":"2017-05-26","proceeding":null,"authors":["Han Zhao","Shanghang Zhang","Guanhang Wu","João P. Costeira","José M. F. Moura","Geoffrey J. Gordon"],"abstract":"While domain adaptation has been actively researched in recent years, most\ntheoretical results and algorithms focus on the single-source-single-target\nadaptation setting. Naive application of such algorithms on multiple source\ndomain adaptation problem may lead to suboptimal solutions. As a step toward\nbridging the gap, we propose a new generalization bound for domain adaptation\nwhen there are multiple source domains with labeled instances and one target\ndomain with unlabeled instances. Compared with existing bounds, the new bound\ndoes not require expert knowledge about the target distribution, nor the\noptimal combination rule for multisource domains. Interestingly, our theory\nalso leads to an efficient learning strategy using adversarial neural networks:\nwe show how to interpret it as learning feature representations that are\ninvariant to the multiple domain shifts while still being discriminative for\nthe learning task. To this end, we propose two models, both of which we call\nmultisource domain adversarial networks (MDANs): the first model optimizes\ndirectly our bound, while the second model is a smoothed approximation of the\nfirst one, leading to a more data-efficient and task-adaptive model. The\noptimization tasks of both models are minimax saddle point problems that can be\noptimized by adversarial training. To demonstrate the effectiveness of MDANs,\nwe conduct extensive experiments showing superior adaptation performance on\nthree real-world datasets: sentiment analysis, digit classification, and\nvehicle counting.","url_abs":"http://arxiv.org/abs/1705.09684v2","url_pdf":"http://arxiv.org/pdf/1705.09684v2.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":"multiple-source-domain-adaptation-with","repo_url":"https://github.com/hanzhaoml/mdan","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"multiple-source-domain-adaptation-with","repo_url":"https://github.com/SanaAwan5/UFDA1","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"multiple-source-domain-adaptation-with","repo_url":"https://github.com/daoyuan98/MSDA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"multiple-source-domain-adaptation-with","repo_url":"https://github.com/dingkmC/MSDA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.09684","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}