Skip to main content

Abstract

Based on the current progress of automated verification for deep learning, it will be a promising research direction to explore the verification of distributed systems where some (or all) of the nodes are implemented with learning models. Such distributed systems present several challenges. First, the communication between nodes and the collaborative learning suggest that existing verification methods will not be directly applicable. Second, such distributed systems may be subject to different types of attacks and therefore require the verification algorithm to consider security and privacy properties other than robustness.

Xiaowei Huang is Professor of Computer Science and Director of the Trustworthy Autonomous Cyber-Physical Systems lab, at the University of Liverpool (UoL). His research is concerned with the development of automated verification techniques that ensure the correctness and reliability of intelligent systems, and he is leading the research direction on the verification and validation of deep neural networks. He co-chairs the AAAI and IJCAI workshop series on Artificial Intelligence Safety.

Search