Explainability and Robustness of Deep Visual Classification Models

TL;DR

This paper investigates the explainability and robustness of deep visual classification models, including CNNs, CapsNets, and ViTs, focusing on their core components to address limitations in interpretability and adversarial vulnerability.

Contribution

The study provides a comparative analysis of the core modules of CNNs, CapsNets, and ViTs to enhance understanding of their explainability and robustness.

Findings

CapsNets and ViTs exhibit different vulnerabilities to adversarial attacks.

Core modules like dynamic routing and self-attention influence model robustness.

Insights into model explainability can guide the development of more transparent models.

Abstract

In the computer vision community, Convolutional Neural Networks (CNNs), first proposed in the 1980's, have become the standard visual classification model. Recently, as alternatives to CNNs, Capsule Networks (CapsNets) and Vision Transformers (ViTs) have been proposed. CapsNets, which were inspired by the information processing of the human brain, are considered to have more inductive bias than CNNs, whereas ViTs are considered to have less inductive bias than CNNs. All three classification models have received great attention since they can serve as backbones for various downstream tasks. However, these models are far from being perfect. As pointed out by the community, there are two weaknesses in standard Deep Neural Networks (DNNs). One of the limitations of DNNs is the lack of explainability. Even though they can achieve or surpass human expert performance in the image classification task, the DNN-based decisions are difficult to understand. In many real-world applications, however, individual decisions need to be explained. The other limitation of DNNs is adversarial vulnerability. Concretely, the small and imperceptible perturbations of inputs can mislead DNNs. The vulnerability of deep neural networks poses challenges to current visual classification models. The potential threats thereof can lead to unacceptable consequences. Besides, studying model adversarial vulnerability can lead to a better understanding of the underlying models. Our research aims to address the two limitations of DNNs. Specifically, we focus on deep visual classification models, especially the core building parts of each classification model, e.g. dynamic routing in CapsNets and self-attention module in ViTs.