{"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/transductive-adversarial-networks-tan","title":"Transductive Adversarial Networks (TAN)","arxiv_id":"1802.02798","date":"2018-02-08","proceeding":null,"authors":["Sean Rowan"],"abstract":"Transductive Adversarial Networks (TAN) is a novel domain-adaptation machine\nlearning framework that is designed for learning a conditional probability\ndistribution on unlabelled input data in a target domain, while also only\nhaving access to: (1) easily obtained labelled data from a related source\ndomain, which may have a different conditional probability distribution than\nthe target domain, and (2) a marginalised prior distribution on the labels for\nthe target domain. TAN leverages a fully adversarial training procedure and a\nunique generator/encoder architecture which approximates the transductive\ncombination of the available source- and target-domain data. A benefit of TAN\nis that it allows the distance between the source- and target-domain\nlabel-vector marginal probability distributions to be greater than 0 (i.e.\ndifferent tasks across the source and target domains) whereas other\ndomain-adaptation algorithms require this distance to equal 0 (i.e. a single\ntask across the source and target domains). TAN can, however, still handle the\nlatter case and is a more generalised approach to this case. Another benefit of\nTAN is that due to being a fully adversarial algorithm, it has the potential to\naccurately approximate highly complex distributions. Theoretical analysis\ndemonstrates the viability of the TAN framework.","url_abs":"http://arxiv.org/abs/1802.02798v1","url_pdf":"http://arxiv.org/pdf/1802.02798v1.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":"transductive-adversarial-networks-tan","repo_url":"https://github.com/sean-rowan/tan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"transductive-adversarial-networks-tan","repo_url":"https://github.com/seanrowan/tan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}