arXiv research

This work optimizes mean estimation under varying privacy constraints.

Introduces six levels of privacy for financial synthetic data.

We consider the problem of reinforcing federated learning with formal privacy guarantees. We propose to employ Bayesian differential privacy, a relaxation of differential privacy for similarly distributed data, to provide sharper privacy loss bounds. We adapt the Bayesian privacy accounting method to the federated sett…

This paper analyzes user-level local differential privacy in distributed systems.

2020 Census uses more noise to protect privacy than needed, improving data accuracy.

GRAND ensures node-level differential privacy for network data.

New bounds for LDP with heterogeneous privacy levels guaranteeing high probability of accuracy.

FLORAS uses orthogonal sequences for SISO FL, offering both DP and convergence guarantees.

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

New algorithms protect user-level privacy in learning tasks.

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

Novel mean estimation method under user-level differential privacy reduces noise in continual mean estimates.

Paper studies user-level differential privacy in federated linear contextual bandits.

Develops privacy-preserving methods for longitudinal linear regression.

Estimates population mean from user-level data with privacy, accounting for heterogeneity.

New method calibrates noise for attack risk, improving ML model accuracy.

This paper sharpens privacy guarantees for high-dimensional PCA under differential privacy.

New privacy-preserving method for conformal prediction without splitting data.

Differential Privacy (DP) provides strong guarantees on the risk of compromising a user's data in statistical learning applications, though these strong protections make learning challenging and may be too stringent for some use cases. To address this, we propose element level differential privacy, which extends differ…

Study improves privacy in cross-silo federated learning by personalizing data.

This paper benchmarks privacy-preserving machine learning on medical images.

The paper investigates how data imbalance affects fairness and accuracy in differentially private deep learning.

New federated $f$-differential privacy for collaborative learning.

We derive the optimal differential privacy (DP) parameters of a mechanism that satisfies a given level of Rényi differential privacy (RDP). Our result is based on the joint range of two $f$-divergences that underlie the approximate and the Rényi variations of differential privacy. We apply our result to the moments acc…

New algorithm optimizes privacy and utility in multi-task learning with skewed data.

FedIRT enables privacy-preserving psychometric estimation without centralizing data.

Iteratively reweighted least squares (IRLS) is a widely-used method in machine learning to estimate the parameters in the generalised linear models. In particular, IRLS for L1 minimisation under the linear model provides a closed-form solution in each step, which is a simple multiplication between the inverse of the we…

We consider the problem of Bayesian learning on sensitive datasets and present two simple but somewhat surprising results that connect Bayesian learning to "differential privacy:, a cryptographic approach to protect individual-level privacy while permiting database-level utility. Specifically, we show that that under s…

A new algorithm estimates mean under varying user data sizes with local differential privacy.

This work optimizes mean estimation under varying user privacy demands.

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 …

The Sampled Gaussian Mechanism's noise level decreases with larger subsampling rates, improving privacy-utility trade-offs.

AdaPrivate-TS: A differentially private Thompson Sampling algorithm for contextual bandits

Based on differential privacy (DP) framework, we introduce and unify privacy definitions for the multi-armed bandit algorithms. We represent the framework with a unified graphical model and use it to connect privacy definitions. We derive and contrast lower bounds on the regret of bandit algorithms satisfying these def…

Novel algorithm reduces privacy noise in machine learning.

Differential privacy is a leading protection setting, focused by design on individual privacy. Many applications, in medical / pharmaceutical domains or social networks, rather posit privacy at a group level, a setting we call integral privacy. We aim for the strongest form of privacy: the group size is in particular n…

Paper improves privacy-accuracy balance in federated learning.

New method uses Winsorized mean estimators for privacy-preserving statistics on dependent data.

Private mean estimation with multiple samples requires a certain number of people to maintain privacy.

Differential privacy protects data privacy by adding noise to data.

A new DP approach for Conformal Prediction using quantile search.

This paper compares two loss functions for learning from aggregated responses and introduces an interpolating estimator.

Data processing inequalities link Fisher information to local differential privacy constraints.

Federated learning is a recent advance in privacy protection. In this context, a trusted curator aggregates parameters optimized in decentralized fashion by multiple clients. The resulting model is then distributed back to all clients, ultimately converging to a joint representative model without explicitly having to s…

Private ALS method improves matrix completion with tighter rates and better privacy.

A framework for privacy-preserving DNN pruning and acceleration.

Hypothesis testing plays a central role in statistical inference, and is used in many settings where privacy concerns are paramount. This work answers a basic question about privately testing simple hypotheses: given two distributions $P$ and $Q$, and a privacy level $\varepsilon$, how many i.i.d. samples are needed to…

Privacy-preserving machine learning methods add randomness, leading to varying predictions.