{"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/differentially-private-database-release-via","title":"Differentially Private Database Release via Kernel Mean Embeddings","arxiv_id":"1710.01641","date":"2017-10-04","proceeding":"ICML 2018 7","authors":["Matej Balog","Ilya Tolstikhin","Bernhard Schölkopf"],"abstract":"We lay theoretical foundations for new database release mechanisms that allow\nthird-parties to construct consistent estimators of population statistics,\nwhile ensuring that the privacy of each individual contributing to the database\nis protected. The proposed framework rests on two main ideas. First, releasing\n(an estimate of) the kernel mean embedding of the data generating random\nvariable instead of the database itself still allows third-parties to construct\nconsistent estimators of a wide class of population statistics. Second, the\nalgorithm can satisfy the definition of differential privacy by basing the\nreleased kernel mean embedding on entirely synthetic data points, while\ncontrolling accuracy through the metric available in a Reproducing Kernel\nHilbert Space. We describe two instantiations of the proposed framework,\nsuitable under different scenarios, and prove theoretical results guaranteeing\ndifferential privacy of the resulting algorithms and the consistency of\nestimators constructed from their outputs.","url_abs":"http://arxiv.org/abs/1710.01641v2","url_pdf":"http://arxiv.org/pdf/1710.01641v2.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":"differentially-private-database-release-via","repo_url":"https://github.com/matejbalog/RKHS-private-database","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}