Detecting Operational Adversarial Examples for Reliable Deep Learning



Zhao, Xingyu ORCID: 0000-0002-3474-349X, Huang, Wei, Schewe, Sven ORCID: 0000-0002-9093-9518, Dong, Yi ORCID: 0000-0003-3047-7777 and Huang, Xiaowei ORCID: 0000-0001-6267-0366
(2021) Detecting Operational Adversarial Examples for Reliable Deep Learning. In: 2021 51st Annual IEEE/IFIP International Conference on Dependable Systems and Networks - Supplemental Volume (DSN-S), 2021-6-21 - 2021-6-24, Taipei, Taiwan.

[img] Text
DSN21_fast_abstract.pdf - Author Accepted Manuscript

Download (721kB) | Preview

Abstract

The utilisation of Deep Learning (DL) raises new challenges regarding its dependability in critical applications. Sound verification and validation methods are needed to assure the safe and reliable use of DL. However, state-of-the-art debug testing methods on DL that aim at detecting adversarial examples (AEs) ignore the operational profile, which statistically depicts the software's future operational use. This may lead to very modest effectiveness on improving the software's delivered reliability, as the testing budget is likely to be wasted on detecting AEs that are unrealistic or encountered very rarely in real-life operation. In this paper, we first present the novel notion of "operational AEs" which are AEs that have relatively high chance to be seen in future operation. Then an initial design of a new DL testing method to efficiently detect "operational AEs" is provided, as well as some insights on our prospective research plan.

Item Type: Conference or Workshop Item (Unspecified)
Additional Information: Preprint accepted by the fast abstract track of DSN'21
Uncontrolled Keywords: Deep Learning robustness, operational profile, safe AI, robustness testing, software reliability, software testing
Divisions: Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science
Depositing User: Symplectic Admin
Date Deposited: 13 Apr 2021 08:20
Last Modified: 15 Mar 2024 11:17
DOI: 10.1109/DSN-S52858.2021.00013
Related URLs:
URI: https://livrepository.liverpool.ac.uk/id/eprint/3119139