activity
20162026
most citedDetection of Adversarial Training Examples in Poisoning Attacks through Anomaly Detection

114 citations · 132 across the 12 of their papers we have counts for

collaborators
Showing 2026Show all

1 paper · 1 filter

cs.LG2026

Adversarial Training with Large Step Sizes: Implicit Bias, Optimization, and Sharpness

Yi Feng, Andrea Paudice

Adversarial Training (AT) is an important method for improving model robustness. We study the dynamics of deterministic Gradient Descent (GD) in AT. Most existing works in this dir…