中文
相关论文

相关论文: On the Geometry of Differential Privacy

200 篇论文

Noise perturbation is one of the most fundamental approaches for achieving $(\epsilon,\delta)$-differential privacy (DP) guarantees when releasing the result of a query or function $f(\cdot)\in\mathbb{R}^M$ evaluated on a sensitive dataset…

密码学与安全 · 计算机科学 2025-12-30 Shuainan Liu , Tianxi Ji , Zhongshuo Fang , Lu Wei , Pan Li

We propose a novel mechanism for answering sets of count- ing queries under differential privacy. Given a workload of counting queries, the mechanism automatically selects a different set of "strategy" queries to answer privately, using…

数据库 · 计算机科学 2012-02-20 Chao Li , Gerome Miklau

The application of Differential Privacy to Natural Language Processing techniques has emerged in relevance in recent years, with an increasing number of studies published in established NLP outlets. In particular, the adaptation of…

计算与语言 · 计算机科学 2024-04-05 Stephen Meisenbacher , Nihildev Nandakumar , Alexandra Klymenko , Florian Matthes

Integer data is typically made differentially private by adding noise from a Discrete Laplace (or Discrete Gaussian) distribution. We study the setting where differential privacy of a counting query is achieved using bit-wise randomized…

信息论 · 计算机科学 2023-11-17 David Rasmussen Lolck , Rasmus Pagh

Datasets are often used multiple times and each successive analysis may depend on the outcome of previous analyses. Standard techniques for ensuring generalization and statistical validity do not account for this adaptive dependence. A…

机器学习 · 计算机科学 2018-06-13 Vitaly Feldman , Thomas Steinke

Differential Privacy (DP) considers a scenario in which an adversary has almost complete information about the entries of a database. This worst-case assumption is likely to overestimate the privacy threat faced by an individual in…

密码学与安全 · 计算机科学 2026-02-11 Dennis Breutigam , Rüdiger Reischuk

We initiate an investigation of node differential privacy for graphs in the local model of private data analysis. In our model, dubbed LNDP*, each node sees its own edge list and releases the output of a local randomizer on this input.…

数据结构与算法 · 计算机科学 2026-04-03 Sofya Raskhodnikova , Adam Smith , Connor Wagaman , Anatoly Zavyalov

The Gaussian Mechanism (GM), which consists in adding Gaussian noise to a vector-valued query before releasing it, is a standard privacy protection mechanism. In particular, given that the query respects some L2 sensitivity property (the L2…

机器学习 · 计算机科学 2024-03-20 Hadrien Hendrikx , Paul Mangold , Aurélien Bellet

We revisit Wald's celebrated Sequential Probability Ratio Test for sequential tests of two simple hypotheses, under privacy constraints. We propose DP-SPRT, a wrapper that can be calibrated to achieve desired error probabilities and privacy…

机器学习 · 统计学 2026-02-05 Thomas Michel , Debabrota Basu , Emilie Kaufmann

We apply the discrepancy method and a chaining approach to give improved bounds on the coreset complexity of a wide class of kernel functions. Our results give randomized polynomial time algorithms to produce coresets of size…

机器学习 · 计算机科学 2023-10-13 Rainie Bozzai , Thomas Rothvoss

Differential private optimization for nonconvex smooth objective is considered. In the previous work, the best known utility bound is $\widetilde O(\sqrt{d}/(n\varepsilon_\mathrm{DP}))$ in terms of the squared full gradient norm, which is…

机器学习 · 计算机科学 2023-06-06 Tomoya Murata , Taiji Suzuki

We study the problem of performing counting queries at different levels in hierarchical structures while preserving individuals' privacy. Motivated by applications, we propose a new error measure for this problem by considering a…

数据结构与算法 · 计算机科学 2023-04-28 Badih Ghazi , Pritish Kamath , Ravi Kumar , Pasin Manurangsi , Kewen Wu

We consider three different variants of differential privacy (DP), namely approximate DP, R\'enyi DP (RDP), and hypothesis test DP. In the first part, we develop a machinery for optimally relating approximate DP to RDP based on the joint…

信息论 · 计算机科学 2021-01-26 Shahab Asoodeh , Jiachun Liao , Flavio P. Calmon , Oliver Kosut , Lalitha Sankar

The collection of individuals' data has become commonplace in many industries. Local differential privacy (LDP) offers a rigorous approach to preserving privacy whereby the individual privatises their data locally, allowing only their…

机器学习 · 计算机科学 2022-05-17 Alex Mansbridge , Gregory Barbour , Davide Piras , Michael Murray , Christopher Frye , Ilya Feige , David Barber

We study the application of differential privacy in hyper-parameter tuning, a crucial process in machine learning involving selecting the best hyper-parameter from several candidates. Unlike many private learning algorithms, including the…

机器学习 · 计算机科学 2025-08-26 Zihang Xiang , Tianhao Wang , Chenglong Wang , Di Wang

In recent years, Gaussian noise has become a popular tool in differentially private algorithms, often replacing Laplace noise which dominated the early literature. Gaussian noise is the standard approach to $\textit{approximate}$…

密码学与安全 · 计算机科学 2024-11-19 Joel Daniel Andersson , Rasmus Pagh , Teresa Anna Steiner , Sahel Torkamani

We consider a dataset $S$ held by an agency, and a vector query of interest, $f(S) \in \mathbb{R}^k$, to be posed by an analyst, which contains the information required for certain planned statistical inference. The agency releases the…

密码学与安全 · 计算机科学 2021-10-13 Tomer Shoham , Yosef Rinott

While preserving the privacy of federated learning (FL), differential privacy (DP) inevitably degrades the utility (i.e., accuracy) of FL due to model perturbations caused by DP noise added to model updates. Existing studies have considered…

机器学习 · 计算机科学 2023-03-09 Xin Yuan , Wei Ni , Ming Ding , Kang Wei , Jun Li , H. Vincent Poor

Prior work on differential privacy analysis of randomized SGD algorithms relies on composition theorems, where the implicit (unrealistic) assumption is that the internal state of the iterative algorithm is revealed to the adversary. As a…

机器学习 · 统计学 2022-10-18 Jiayuan Ye , Reza Shokri

The Gaussian mechanism is one differential privacy mechanism commonly used to protect numerical data. However, it may be ill-suited to some applications because it has unbounded support and thus can produce invalid numerical answers to…

密码学与安全 · 计算机科学 2022-12-01 Bo Chen , Matthew Hale