Towards Class-wise Robustness Analysis

TL;DR

This paper investigates class-wise robustness in deep neural networks, revealing that class vulnerabilities are influenced by false positive tendencies and highlighting the importance of class-specific analysis for improving model robustness.

Contribution

It introduces a class-wise robustness analysis framework, focusing on latent space structures and false positive impacts, to better understand and evaluate class-specific vulnerabilities.

Findings

False positives significantly affect class vulnerability.

Class-wise robustness varies independently of overall accuracy.

Class False Positive Score effectively measures class susceptibility.

Abstract

While being very successful in solving many downstream tasks, the application of deep neural networks is limited in real-life scenarios because of their susceptibility to domain shifts such as common corruptions, and adversarial attacks. The existence of adversarial examples and data corruption significantly reduces the performance of deep classification models. Researchers have made strides in developing robust neural architectures to bolster decisions of deep classifiers. However, most of these works rely on effective adversarial training methods, and predominantly focus on overall model robustness, disregarding class-wise differences in robustness, which are critical. Exploiting weakly robust classes is a potential avenue for attackers to fool the image recognition models. Therefore, this study investigates class-to-class biases across adversarially trained robust classification models to understand their latent space structures and analyze their strong and weak class-wise properties. We further assess the robustness of classes against common corruptions and adversarial attacks, recognizing that class vulnerability extends beyond the number of correct classifications for a specific class. We find that the number of false positives of classes as specific target classes significantly impacts their vulnerability to attacks. Through our analysis on the Class False Positive Score, we assess a fair evaluation of how susceptible each class is to misclassification.