arXiv research

Unified framework for subsampling mechanisms with tighter privacy guarantees.

The paper improves privacy accounting for discrete-valued mechanisms and the subsampled Gaussian mechanism.

Improves privacy guarantees by analyzing randomness in privacy-preserving mechanisms.

New insights into privacy guarantees for subsampled mechanisms under composition.

New mechanism protects neural network weights from privacy attacks during self-supervised learning.

A new discrete privacy mechanism for federated learning.

Novel compression method preserves privacy while reducing communication costs.

New method compares DP mechanisms beyond $(\varepsilon, δ)$ pairs.

Many modern databases include personal and sensitive correlated data, such as private information on users connected together in a social network, and measurements of physical activity of single subjects across time. However, differential privacy, the current gold standard in data privacy, does not adequately address p…

A new Gaussian mechanism for differential privacy in the shuffle model is introduced.

New mechanisms improve differential privacy for scalar queries.

Paper develops novel privacy mechanism for Riemannian manifold data using geometric analysis and heat diffusion.

Expands differential privacy mechanisms to include the Generalized Gaussian mechanism for improved private machine learning.

New privacy mechanism reduces error in query results.

Combines public and private data for better statistical estimation.

Enhances privacy in machine learning through Rényi Pufferfish mechanisms.

Conformal-DP improves differential privacy on manifold data by calibrating perturbations based on local densities.

New methods test discrete distributions faster with local privacy constraints.

New privacy mechanism for user-level discrete distributions with reduced penalty.

Proposes a privacy-preserving recommendation system using matrix factorization and differential privacy.

The collection and analysis of user data drives improvements in the app and web ecosystems, but comes with risks to privacy. This paper examines discrete distribution estimation under local privacy, a setting wherein service providers can learn the distribution of a categorical statistic of interest without collecting …

New privacy method for eye tracking data reduces correlations and maintains accuracy.

Differential privacy mechanism design has traditionally been tailored for a scalar-valued query function. Although many mechanisms such as the Laplace and Gaussian mechanisms can be extended to a matrix-valued query function by adding i.i.d. noise to each element of the matrix, this method is often suboptimal as it for…

Interactive privacy mechanisms improve spectral density estimation under local differential privacy.

PBM mechanism improves privacy and accuracy in federated learning.

A fundamental result in differential privacy states that the privacy guarantees of a mechanism are preserved by any post-processing of its output. In this paper we investigate under what conditions stochastic post-processing can amplify the privacy of a mechanism. By interpreting post-processing as the application of a…

The Sampled Gaussian Mechanism (SGM)---a composition of subsampling and the additive Gaussian noise---has been successfully used in a number of machine learning applications. The mechanism's unexpected power is derived from privacy amplification by sampling where the privacy cost of a single evaluation diminishes quadr…

Study privacy and accuracy in high-dimensional LASSO with perturbation mechanisms.

A new mechanism for differentially private Fréchet mean on SPD matrices.

New mechanism for pure differential privacy on functional summaries using Laplace-like process.

Optimal DP mechanisms for vector queries are found to be staircase distributions.

Designing a data sharing mechanism without sacrificing too much privacy can be considered as a game between data holders and malicious attackers. This paper describes a compressive adversarial privacy framework that captures the trade-off between the data privacy and utility. We characterize the optimal data releasing …

Develops new instance-optimality concepts in differential privacy.

Paper develops efficient mechanisms for estimating variance and covariance under differential privacy in the add-remove model.

Paper improves privacy bounds for shuffle model using novel numerical techniques.

With the proliferation of mobile devices and the internet of things, developing principled solutions for privacy in time series applications has become increasingly important. While differential privacy is the gold standard for database privacy, many time series applications require a different kind of guarantee, and a…

Optimal LDP mechanisms reduce data unfairness in classification.

New algorithm maintains privacy while improving model performance in selective release.

In the post-industrial world, data science and analytics have gained paramount importance regarding digital data privacy. Improper methods of establishing privacy for accessible datasets can compromise large amounts of user data even if the adversary has a small amount of preliminary knowledge of a user. Many researche…

Differential privacy is a strong notion for privacy that can be used to prove formal guarantees, in terms of a privacy budget, $ε$, about how much information is leaked by a mechanism. However, implementations of privacy-preserving machine learning often select large values of $ε$ in order to get acceptable utility of …

A new privacy accountant for Gaussian differential privacy measures individual privacy losses.

Optimal Gaussian noise mechanisms achieve nearly optimal error in unbiased mean estimation.

New approach preserves privacy in high-dimensional data using representation learning.

The Dirichlet mechanism protects privacy while minimizing KL divergence.

New DP mechanism SWAG-PPM improves privacy in deep learning models.

We present a data-driven framework called generative adversarial privacy (GAP). Inspired by recent advancements in generative adversarial networks (GANs), GAP allows the data holder to learn the privatization mechanism directly from the data. Under GAP, finding the optimal privacy mechanism is formulated as a constrain…

LDP is equivalent to contraction of E_γ-divergence, impacting privacy and utility.

Differentially private log-location-scale regression models improve privacy in statistical analysis.