1

On the Learnability of Distribution Classes with Adaptive Adversaries

Optimal Differentially Private Sampling of Unbounded Gaussians

Avoiding Pitfalls for Privacy Accounting of Subsampled Mechanisms under Composition

Choosing Public Datasets for Private Machine Learning via Gradient Subspace Distance

Machine Unlearning Fails to Remove Data Poisoning Attacks

Position: Membership Inference Attacks Cannot Prove that a Model Was Trained On Your Data

Private Mean Estimation with Person-Level Differential Privacy

Differentially Private Post-Processing for Fair Regression

Disguised Copyright Infringement of Latent Diffusion Models

Position: Considerations for Differentially Private Learning with Large-Scale Public Pretraining