{"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/generalizing-to-unseen-domains-via","title":"Generalizing to Unseen Domains via Adversarial Data Augmentation","arxiv_id":"1805.12018","date":"2018-05-30","proceeding":"NeurIPS 2018 12","authors":["Riccardo Volpi","Hongseok Namkoong","Ozan Sener","John Duchi","Vittorio Murino","Silvio Savarese"],"abstract":"We are concerned with learning models that generalize well to different\n\\emph{unseen} domains. We consider a worst-case formulation over data\ndistributions that are near the source domain in the feature space. Only using\ntraining data from a single source distribution, we propose an iterative\nprocedure that augments the dataset with examples from a fictitious target\ndomain that is \"hard\" under the current model. We show that our iterative\nscheme is an adaptive data augmentation method where we append adversarial\nexamples at each iteration. For softmax losses, we show that our method is a\ndata-dependent regularization scheme that behaves differently from classical\nregularizers that regularize towards zero (e.g., ridge or lasso). On digit\nrecognition and semantic segmentation tasks, our method learns models improve\nperformance across a range of a priori unknown target domains.","url_abs":"http://arxiv.org/abs/1805.12018v2","url_pdf":"http://arxiv.org/pdf/1805.12018v2.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":"generalizing-to-unseen-domains-via","repo_url":"https://github.com/ricvolpi/generalize-unseen-domains","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.12018","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}