Sharpness-Aware Initialization: Improving Differentially Private Machine Learning from First Principles

Name: Sharpness-Aware Initialization: Improving Differentially Private Machine Learning from First Principles
Duration: 15 min
Description: Talk from 34th USENIX Security Symposium (USENIX Security '25).

Zihao Wang

34th USENIX Security Symposium (USENIX Security '25) · Day 2 · ML and AI Privacy 1: Federated Learning and Protecting Data