Updated Results for Confidence-Calibrated Adversarial Training

Since I worked on confidence-calibrated training (CCAT) some years ago, CCAT has been evaluated using novel attacks. In this article, I want to share some updated results and numbers and contrast the reported numbers with newer experiments that I ran.

The post Updated Results for Confidence-Calibrated Adversarial Training appeared first on David Stutz.

adversarial machine learning article attacks blog confidence david deep learning machine learning novel numbers training