On the Over-Memorization During Natural, Robust and Catastrophic Overfitting

arXiv:2310.08847v2 Announce Type: replace
Abstract: Overfitting negatively impacts the generalization ability of deep neural networks (DNNs) in both natural and adversarial training. Existing methods struggle to consistently address different types of overfitting, typically designing strategies that focus separately on either natural or adversarial patterns. In this work, we adopt a unified perspective by solely focusing on natural patterns to explore different types of overfitting. Specifically, we examine the memorization effect in DNNs and reveal a shared behaviour termed over-memorization, which …

abstract adversarial adversarial training arxiv cs.lg designing focus impacts natural networks neural networks overfitting patterns robust strategies struggle training type types work