A Survey of Safety and Trustworthiness of Deep Neural Networks: Verification, Testing, Adversarial Attack and Defence, and Interpretability

TL;DR

This survey reviews recent research on making deep neural networks safer and more trustworthy, focusing on verification, testing, adversarial defenses, and interpretability, covering 202 papers mostly published after 2017.

Contribution

It provides a comprehensive overview of current methods and progress in ensuring DNN safety and trustworthiness across four key areas.

Findings

Significant progress in DNN verification and testing methods.

Advances in adversarial attack detection and defense strategies.

Growing emphasis on interpretability for trustworthiness.

Abstract

In the past few years, significant progress has been made on deep neural networks (DNNs) in achieving human-level performance on several long-standing tasks. With the broader deployment of DNNs on various applications, the concerns over their safety and trustworthiness have been raised in public, especially after the widely reported fatal incidents involving self-driving cars. Research to address these concerns is particularly active, with a significant number of papers released in the past few years. This survey paper conducts a review of the current research effort into making DNNs safe and trustworthy, by focusing on four aspects: verification, testing, adversarial attack and defence, and interpretability. In total, we survey 202 papers, most of which were published after 2017.