{"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/the-variational-fair-autoencoder","title":"The Variational Fair Autoencoder","arxiv_id":"1511.00830","date":"2015-11-03","proceeding":null,"authors":["Christos Louizos","Kevin Swersky","Yujia Li","Max Welling","Richard Zemel"],"abstract":"We investigate the problem of learning representations that are invariant to\ncertain nuisance or sensitive factors of variation in the data while retaining\nas much of the remaining information as possible. Our model is based on a\nvariational autoencoding architecture with priors that encourage independence\nbetween sensitive and latent factors of variation. Any subsequent processing,\nsuch as classification, can then be performed on this purged latent\nrepresentation. To remove any remaining dependencies we incorporate an\nadditional penalty term based on the \"Maximum Mean Discrepancy\" (MMD) measure.\nWe discuss how these architectures can be efficiently trained on data and show\nin experiments that this method is more effective than previous work in\nremoving unwanted sources of variation while maintaining informative latent\nrepresentations.","url_abs":"http://arxiv.org/abs/1511.00830v6","url_pdf":"http://arxiv.org/pdf/1511.00830v6.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":"the-variational-fair-autoencoder","repo_url":"https://github.com/nctumllab/huang-ching-wei","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"the-variational-fair-autoencoder","repo_url":"https://github.com/yevgeni-integrate-ai/vfae","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/sentiment-analysis-on-multi-domain-sentiment","task":"Sentiment Analysis","dataset":"Multi-Domain Sentiment Dataset","model":"VFAE","rank_in_archive_order":4,"of":6,"metrics":{"Average":"78.36","Books":"73.40","DVD":"76.57","Electronics":"80.53","Kitchen":"82.93"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.00830","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}