{"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/learning-anonymized-representations-with","title":"Learning Anonymized Representations with Adversarial Neural Networks","arxiv_id":"1802.09386","date":"2018-02-26","proceeding":null,"authors":["Clément Feutry","Pablo Piantanida","Yoshua Bengio","Pierre Duhamel"],"abstract":"Statistical methods protecting sensitive information or the identity of the\ndata owner have become critical to ensure privacy of individuals as well as of\norganizations. This paper investigates anonymization methods based on\nrepresentation learning and deep neural networks, and motivated by novel\ninformation theoretical bounds. We introduce a novel training objective for\nsimultaneously training a predictor over target variables of interest (the\nregular labels) while preventing an intermediate representation to be\npredictive of the private labels. The architecture is based on three\nsub-networks: one going from input to representation, one from representation\nto predicted regular labels, and one from representation to predicted private\nlabels. The training procedure aims at learning representations that preserve\nthe relevant part of the information (about regular labels) while dismissing\ninformation about the private labels which correspond to the identity of a\nperson. We demonstrate the success of this approach for two distinct\nclassification versus anonymization tasks (handwritten digits and sentiment\nanalysis).","url_abs":"http://arxiv.org/abs/1802.09386v1","url_pdf":"http://arxiv.org/pdf/1802.09386v1.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":"learning-anonymized-representations-with","repo_url":"https://github.com/maxfriedrich/deid-training-data","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.09386","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}