{"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-generalization-via-conditional","title":"Domain Generalization via Conditional Invariant Representation","arxiv_id":"1807.08479","date":"2018-07-23","proceeding":null,"authors":["Ya Li","Mingming Gong","Xinmei Tian","Tongliang Liu","DaCheng Tao"],"abstract":"Domain generalization aims to apply knowledge gained from multiple labeled\nsource domains to unseen target domains. The main difficulty comes from the\ndataset bias: training data and test data have different distributions, and the\ntraining set contains heterogeneous samples from different distributions. Let\n$X$ denote the features, and $Y$ be the class labels. Existing domain\ngeneralization methods address the dataset bias problem by learning a\ndomain-invariant representation $h(X)$ that has the same marginal distribution\n$\\mathbb{P}(h(X))$ across multiple source domains. The functional relationship\nencoded in $\\mathbb{P}(Y|X)$ is usually assumed to be stable across domains\nsuch that $\\mathbb{P}(Y|h(X))$ is also invariant. However, it is unclear\nwhether this assumption holds in practical problems. In this paper, we consider\nthe general situation where both $\\mathbb{P}(X)$ and $\\mathbb{P}(Y|X)$ can\nchange across all domains. We propose to learn a feature representation which\nhas domain-invariant class conditional distributions $\\mathbb{P}(h(X)|Y)$. With\nthe conditional invariant representation, the invariance of the joint\ndistribution $\\mathbb{P}(h(X),Y)$ can be guaranteed if the class prior\n$\\mathbb{P}(Y)$ does not change across training and test domains. Extensive\nexperiments on both synthetic and real data demonstrate the effectiveness of\nthe proposed method.","url_abs":"http://arxiv.org/abs/1807.08479v1","url_pdf":"http://arxiv.org/pdf/1807.08479v1.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-generalization-via-conditional","repo_url":"https://github.com/facebookresearch/DomainBed","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"domain-generalization","task_name":"Domain Generalization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.08479","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}