中文
相关论文

相关论文: R\'enyi Differential Privacy of the Sampled Gaussi…

200 篇论文

While modern machine learning models rely on increasingly large training datasets, data is often limited in privacy-sensitive domains. Generative models trained with differential privacy (DP) on sensitive data can sidestep this challenge,…

机器学习 · 统计学 2024-01-02 Tim Dockhorn , Tianshi Cao , Arash Vahdat , Karsten Kreis

With the rapid growth of digital platforms, there is increasing apprehension about how personal data is being collected, stored, and used by various entities. These concerns range from data breaches and cyber-attacks to potential misuse of…

密码学与安全 · 计算机科学 2023-11-30 Christopher Zawacki , Eyad Abed

Data-dependent privacy accounting frameworks such as per-instance differential privacy (pDP) and Fisher information loss (FIL) confer fine-grained privacy guarantees for individuals in a fixed training dataset. These guarantees can be…

密码学与安全 · 计算机科学 2024-03-12 Shengyuan Hu , Saeed Mahloujifar , Virginia Smith , Kamalika Chaudhuri , Chuan Guo

Individual privacy accounting enables bounding differential privacy (DP) loss individually for each participant involved in the analysis. This can be informative as often the individual privacy losses are considerably smaller than those…

密码学与安全 · 计算机科学 2023-08-25 Antti Koskela , Marlon Tobaben , Antti Honkela

Differential privacy (DP) has become a rigorous central concept for privacy protection in the past decade. We use Gaussian differential privacy (GDP) in gauging the level of privacy protection for releasing statistical summaries from data.…

统计方法学 · 统计学 2025-04-01 Minwoo Kim , Jonghyeok Lee , Seung Woo Kwak , Sungkyu Jung

The randomized power method has gained significant interest due to its simplicity and efficient handling of large-scale spectral analysis and recommendation tasks. However, its application to large datasets containing personal information…

机器学习 · 计算机科学 2025-06-13 Julien Nicolas , César Sabater , Mohamed Maouche , Sonia Ben Mokhtar , Mark Coates

Although distributed Gaussian process regression (GPR) enables multiple agents with separate datasets to jointly learn a model of the target function, its collaborative nature poses risks of private data leakage. To address this, we propose…

系统与控制 · 电气工程与系统科学 2025-12-08 Yeongjun Jang , Kaoru Teranishi , Jihoon Suh , Takashi Tanaka

We study a basic private estimation problem: each of $n$ users draws a single i.i.d. sample from an unknown Gaussian distribution, and the goal is to estimate the mean of this Gaussian distribution while satisfying local differential…

机器学习 · 计算机科学 2019-10-29 Matthew Joseph , Janardhan Kulkarni , Jieming Mao , Zhiwei Steven Wu

A central issue in machine learning is how to train models on sensitive user data. Industry has widely adopted a simple algorithm: Stochastic Gradient Descent with noise (a.k.a. Stochastic Gradient Langevin Dynamics). However, foundational…

机器学习 · 计算机科学 2023-03-01 Jason M. Altschuler , Kunal Talwar

Differentially private selection mechanisms offer strong privacy guarantees for queries aiming to identify the top-scoring element r from a finite set R, based on a dataset-dependent utility function. While selection queries are fundamental…

机器学习 · 计算机科学 2025-04-16 Iago Chaves , Victor Farias , Amanda Perez , Diego Mesquita , Javam Machado

Differentially Private Stochastic Gradient Descent (DP-SGD) is a key method for applying privacy in the training of deep learning models. It applies isotropic Gaussian noise to gradients during training, which can perturb these gradients in…

机器学习 · 计算机科学 2023-11-28 Pedro Faustini , Natasha Fernandes , Shakila Tonni , Annabelle McIver , Mark Dras

We develop a sharp, experiment-level privacy theory for amplification by shuffling in the Gaussian regime: a fixed finite-output local randomizer with full support and neighboring binary datasets differing in one user. We first prove exact…

信息论 · 计算机科学 2026-03-24 Alex Shvets

We show that Thompson Sampling with Gaussian Prior as detailed by Algorithm 2 in (Agrawal & Goyal, 2013) is already differentially private. Theorem 1 show that it enjoys a very competitive privacy loss of only $\mathcal{O}(\ln^2 T)$ after T…

密码学与安全 · 计算机科学 2018-06-26 Aristide C. Y. Tossou , Christos Dimitrakakis

Machine learning (ML) models trained by differentially private stochastic gradient descent (DP-SGD) have much lower utility than the non-private ones. To mitigate this degradation, we propose a DP Laplacian smoothing SGD (DP-LSSGD) to train…

机器学习 · 计算机科学 2019-12-10 Bao Wang , Quanquan Gu , March Boedihardjo , Farzin Barekat , Stanley J. Osher

Differential privacy (DP) is a popular mechanism for training machine learning models with bounded leakage about the presence of specific points in the training data. The cost of differential privacy is a reduction in the model's accuracy.…

机器学习 · 计算机科学 2019-10-29 Eugene Bagdasaryan , Vitaly Shmatikov

We develop an advanced approach for extending Gaussian Differential Privacy (GDP) to general Riemannian manifolds. The concept of GDP stands out as a prominent privacy definition that strongly warrants extension to manifold settings, due to…

密码学与安全 · 计算机科学 2023-11-20 Yangdi Jiang , Xiaotian Chang , Yi Liu , Lei Ding , Linglong Kong , Bei Jiang

Deep learning models are often trained on datasets that contain sensitive information such as individuals' shopping transactions, personal contacts, and medical records. An increasingly important line of work therefore has sought to train…

机器学习 · 计算机科学 2020-07-23 Zhiqi Bu , Jinshuo Dong , Qi Long , Weijie J. Su

Differential privacy provides the first theoretical foundation with provable privacy guarantee against adversaries with arbitrary prior knowledge. The main idea to achieve differential privacy is to inject random noise into statistical…

数据结构与算法 · 计算机科学 2011-11-01 Yang D. Li , Zhenjie Zhang , Marianne Winslett , Yin Yang

Federated Learning (FL) is an emerging paradigm that holds great promise for privacy-preserving machine learning using distributed data. To enhance privacy, FL can be combined with Differential Privacy (DP), which involves adding Gaussian…

机器学习 · 计算机科学 2024-12-10 Tianqu Kang , Lumin Liu , Hengtao He , Jun Zhang , S. H. Song , Khaled B. Letaief

Differential privacy comes equipped with multiple analytical tools for the design of private data analyses. One important tool is the so-called "privacy amplification by subsampling" principle, which ensures that a differentially private…

机器学习 · 计算机科学 2018-11-26 Borja Balle , Gilles Barthe , Marco Gaboardi