{"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/inspectre-privately-estimating-the-unseen","title":"INSPECTRE: Privately Estimating the Unseen","arxiv_id":"1803.00008","date":"2018-02-28","proceeding":"ICML 2018 7","authors":["Jayadev Acharya","Gautam Kamath","Ziteng Sun","Huanyu Zhang"],"abstract":"We develop differentially private methods for estimating various\ndistributional properties. Given a sample from a discrete distribution $p$,\nsome functional $f$, and accuracy and privacy parameters $\\alpha$ and\n$\\varepsilon$, the goal is to estimate $f(p)$ up to accuracy $\\alpha$, while\nmaintaining $\\varepsilon$-differential privacy of the sample.\n  We prove almost-tight bounds on the sample size required for this problem for\nseveral functionals of interest, including support size, support coverage, and\nentropy. We show that the cost of privacy is negligible in a variety of\nsettings, both theoretically and experimentally. Our methods are based on a\nsensitivity analysis of several state-of-the-art methods for estimating these\nproperties with sublinear sample complexities.","url_abs":"http://arxiv.org/abs/1803.00008v1","url_pdf":"http://arxiv.org/pdf/1803.00008v1.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":"inspectre-privately-estimating-the-unseen","repo_url":"https://github.com/HuanyuZhang/INSPECTRE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"sensitivity","task_name":"Sensitivity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1803.00008","atlas_url":"https://app.syntology.ai/?focus=1803.00008","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}