Improving the Accuracy-Robustness Trade-Off of Classifiers via Adaptive Smoothing

arXiv:2301.12554v4 Announce Type: replace
Abstract: While prior research has proposed a plethora of methods that build neural classifiers robust against adversarial robustness, practitioners are still reluctant to adopt them due to their unacceptably severe clean accuracy penalties. This paper significantly alleviates this accuracy-robustness trade-off by mixing the output probabilities of a standard classifier and a robust classifier, where the standard network is optimized for clean accuracy and is not robust in general. We show that the robust base classifier's confidence …

accuracy arxiv classifiers cs.cr cs.cv cs.lg improving robustness trade trade-off type via