10 papers
Benign Overfitting in Adversarial Training for Vision Transformers
Jiaming Zhang, Meng Ding, Shaopeng Fu +2
Despite the remarkable success of Vision Transformers (ViTs) across a wide range of vision tasks, recent studies have revealed that they remain vulnerable to adversarial examples,…
Provable Effects of Data Replay in Continual Learning: A Feature Learning Perspective
Meng Ding, Jinhui Xu, Kaiyi Ji
Continual learning (CL) aims to train models on a sequence of tasks while retaining performance on previously learned ones. A core challenge in this setting is catastrophic forgett…
Finding Differentially Private Second Order Stationary Points in Stochastic Minimax Optimization
Difei Xu, Youming Tao, Meng Ding +2
We provide the first study of the problem of finding differentially private (DP) second-order stationary points (SOSP) in stochastic (non-convex) minimax optimization. Existing lit…
Understanding Fine-tuning in Approximate Unlearning: A Theoretical Perspective
Meng Ding, Rohan Sharma, Changyou Chen +2
Machine Unlearning has emerged as a significant area of research, focusing on `removing' specific subsets of data from a trained model. Fine-tuning (FT) methods have become one of…
Beyond Ordinary Lipschitz Constraints: Differentially Private Stochastic Optimization with Tsybakov Noise Condition
Difei Xu, Meng Ding, Zihang Xiang +2
We study Stochastic Convex Optimization in the Differential Privacy model (DP-SCO). Unlike previous studies, here we assume the population risk function satisfies the Tsybakov Nois…
Nearly Optimal Differentially Private ReLU Regression
Meng Ding, Mingxi Lei, Shaowei Wang +3
In this paper, we investigate one of the most fundamental nonconvex learning problems, ReLU regression, in the Differential Privacy (DP) model. Previous studies on private ReLU reg…