{"language":[{"iso":"eng"}],"publisher":"Neural Information Processing Systems Foundation","day":"20","status":"public","type":"conference","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","conference":{"start_date":"2024-12-09","location":"Vancouver, Canada","end_date":"2024-12-15","name":"NeurIPS: Neural Information Processing Systems"},"external_id":{"arxiv":["2406.03802"]},"OA_place":"repository","corr_author":"1","date_updated":"2025-05-14T11:33:22Z","ec_funded":1,"scopus_import":"1","publication":"38th Conference on Neural Information Processing Systems","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2406.03802"}],"title":"Continual counting with gradual privacy expiration","intvolume":"        37","OA_type":"green","oa":1,"article_processing_charge":"No","quality_controlled":"1","arxiv":1,"author":[{"first_name":"Joel Daniel","last_name":"Andersson","full_name":"Andersson, Joel Daniel"},{"id":"540c9bbd-f2de-11ec-812d-d04a5be85630","first_name":"Monika H","full_name":"Henzinger, Monika H","last_name":"Henzinger","orcid":"0000-0002-5008-6530"},{"last_name":"Pagh","full_name":"Pagh, Rasmus","first_name":"Rasmus"},{"full_name":"Steiner, Teresa Anna","last_name":"Steiner","first_name":"Teresa Anna"},{"first_name":"Jalaj","full_name":"Upadhyay, Jalaj","last_name":"Upadhyay"}],"_id":"19512","volume":37,"acknowledgement":"Monika Henzinger: This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 101019564) and the Austrian Science Fund (FWF) grant DOI 10.55776/Z422, grant DOI 10.55776/I5982, and grant DOI 10.55776/P33775 with additional funding from the netidee SCIENCE Stiftung, 2020–2024. Joel Daniel Andersson and Rasmus Pagh are affiliated with Basic Algorithms Research Copenhagen (BARC), supported by the VILLUM Foundation grant 16582, and are also supported by Providentia, a Data Science Distinguished Investigator grant from Novo Nordisk Fonden. Teresa Anna Steiner is supported by a research grant (VIL51463) from VILLUM FONDEN. This work was done while Teresa Anna Steiner was a Postdoc at the Technical University of Denmark. Jalaj Upadhyay’s research was funded by the Rutgers Decanal Grant no. 302918 and an unrestricted gift from Google.","abstract":[{"lang":"eng","text":"Differential privacy with gradual expiration models the setting where data items\r\narrive in a stream and at a given time t the privacy loss guaranteed for a data item\r\nseen at time (t − d) is εg(d), where g is a monotonically non-decreasing function.\r\nWe study the fundamental continual (binary) counting problem where each data\r\nitem consists of a bit, and the algorithm needs to output at each time step the sum of\r\nall the bits streamed so far. For a stream of length T and privacy without expiration\r\ncontinual counting is possible with maximum (over all time steps) additive error\r\nO(log2\r\n(T)/ε) and the best known lower bound is Ω(log(T)/ε); closing this gap\r\nis a challenging open problem.\r\nWe show that the situation is very different for privacy with gradual expiration by\r\ngiving upper and lower bounds for a large set of expiration functions g. Specifically,\r\nour algorithm achieves an additive error of O(log(T)/ε) for a large set of privacy\r\nexpiration functions. We also give a lower bound that shows that if C is the additive\r\nerror of any ε-DP algorithm for this problem, then the product of C and the privacy\r\nexpiration function after 2C steps must be Ω(log(T)/ε). Our algorithm matches\r\nthis lower bound as its additive error is O(log(T)/ε), even when g(2C) = O(1).\r\nOur empirical evaluation shows that we achieve a slowly growing privacy loss\r\nwith significantly smaller empirical privacy loss for large values of d than a natural\r\nbaseline algorithm."}],"department":[{"_id":"MoHe"}],"publication_identifier":{"issn":["1049-5258"]},"month":"12","date_published":"2024-12-20T00:00:00Z","alternative_title":["Advances in Neural Information Processing Systems"],"oa_version":"Preprint","project":[{"call_identifier":"H2020","grant_number":"101019564","_id":"bd9ca328-d553-11ed-ba76-dc4f890cfe62","name":"The design and evaluation of modern fully dynamic data structures"},{"name":"Efficient algorithms","_id":"34def286-11ca-11ed-8bc3-da5948e1613c","grant_number":"Z00422"},{"name":"Static and Dynamic Hierarchical Graph Decompositions","grant_number":"I05982","_id":"bda196b2-d553-11ed-ba76-8e8ee6c21103"},{"grant_number":"P33775","_id":"bd9e3a2e-d553-11ed-ba76-8aa684ce17fe","name":"Fast Algorithms for a Reactive Network Layer"}],"date_created":"2025-04-06T22:01:32Z","year":"2024","citation":{"ista":"Andersson JD, Henzinger M, Pagh R, Steiner TA, Upadhyay J. 2024. Continual counting with gradual privacy expiration. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","mla":"Andersson, Joel Daniel, et al. “Continual Counting with Gradual Privacy Expiration.” <i>38th Conference on Neural Information Processing Systems</i>, vol. 37, Neural Information Processing Systems Foundation, 2024.","apa":"Andersson, J. D., Henzinger, M., Pagh, R., Steiner, T. A., &#38; Upadhyay, J. (2024). Continual counting with gradual privacy expiration. In <i>38th Conference on Neural Information Processing Systems</i> (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ama":"Andersson JD, Henzinger M, Pagh R, Steiner TA, Upadhyay J. Continual counting with gradual privacy expiration. In: <i>38th Conference on Neural Information Processing Systems</i>. Vol 37. Neural Information Processing Systems Foundation; 2024.","chicago":"Andersson, Joel Daniel, Monika Henzinger, Rasmus Pagh, Teresa Anna Steiner, and Jalaj Upadhyay. “Continual Counting with Gradual Privacy Expiration.” In <i>38th Conference on Neural Information Processing Systems</i>, Vol. 37. Neural Information Processing Systems Foundation, 2024.","ieee":"J. D. Andersson, M. Henzinger, R. Pagh, T. A. Steiner, and J. Upadhyay, “Continual counting with gradual privacy expiration,” in <i>38th Conference on Neural Information Processing Systems</i>, Vancouver, Canada, 2024, vol. 37.","short":"J.D. Andersson, M. Henzinger, R. Pagh, T.A. Steiner, J. Upadhyay, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024."},"publication_status":"published"}