A survey of safety and trustworthiness of deep neural networks: Verification, testing, adversarial attack and defence, and interpretability?

Huang, Xiaowei ORCID: 0000-0001-6267-0366, Kroening, Daniel, Ruan, Wenjie, Sharp, James, Sun, Youcheng, Thamo, Emese, Wu, Min and Yi, Xinping ORCID: 0000-0001-5163-2364 (2020) A survey of safety and trustworthiness of deep neural networks: Verification, testing, adversarial attack and defence, and interpretability? COMPUTER SCIENCE REVIEW, 37. p. 100270.

Text 1812.08342v1.pdf - Submitted version Download (1MB)

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.

Item Type:	Article
Additional Information:	To appear in the journal of Computer Science Review
Uncontrolled Keywords:	cs.LG, cs.LG, cs.AI, I.2; F.3.1
Depositing User:	Symplectic Admin
Date Deposited:	07 Jan 2019 10:52
Last Modified:	19 Jan 2023 01:08
DOI:	10.1016/j.cosrev.2020.100270
Related URLs:	Author Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3030551