{"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/domain-adaptation-with-randomized-expectation","title":"Domain Adaptation with Randomized Expectation Maximization","arxiv_id":"1803.07634","date":"2018-03-20","proceeding":null,"authors":["Twan van Laarhoven","Elena Marchiori"],"abstract":"Domain adaptation (DA) is the task of classifying an unlabeled dataset\n(target) using a labeled dataset (source) from a related domain. The majority\nof successful DA methods try to directly match the distributions of the source\nand target data by transforming the feature space. Despite their success, state\nof the art methods based on this approach are either involved or unable to\ndirectly scale to data with many features. This article shows that domain\nadaptation can be successfully performed by using a very simple randomized\nexpectation maximization (EM) method. We consider two instances of the method,\nwhich involve logistic regression and support vector machine, respectively. The\nunderlying assumption of the proposed method is the existence of a good single\nlinear classifier for both source and target domain. The potential limitations\nof this assumption are alleviated by the flexibility of the method, which can\ndirectly incorporate deep features extracted from a pre-trained deep neural\nnetwork. The resulting algorithm is strikingly easy to implement and apply. We\ntest its performance on 36 real-life adaptation tasks over text and image data\nwith diverse characteristics. The method achieves state-of-the-art results,\ncompetitive with those of involved end-to-end deep transfer-learning methods.","url_abs":"http://arxiv.org/abs/1803.07634v1","url_pdf":"http://arxiv.org/pdf/1803.07634v1.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":"domain-adaptation-with-randomized-expectation","repo_url":"https://github.com/twanvl/adrem","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[{"method_slug":"logistic-regression","method_name":"Logistic Regression"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}