Neural Network Testing, Verification & Repair
Neural networks can be (arguably) viewed a different paradigm of programming, where logical reasoning is replaced with big data and optimization. Unlike traditional programs, however, neural networks are subject to bugs, e.g., adversarial samples and discriminatory instances. In this line of work, we aim to develop systematic theories, methods and tools to improve the quality of AI-systems.
Key Publications
Causality-Based Neural Network Repair.
44th International Conference on Software Engineering, ICSE 2022, Pittsburgh, USA, 2022..
Probabilistic Verification of Neural Networks Against Group Fairness.
24th International Symposium on Formal Methods, FM 2021, Beijng, China, 20-26 November, 2021..
Improving Neural Network Verification through Spurious Region Guided Refinement.
International Conference on Tools and Algorithms for the Construction and Analysis of Systems, TACAS 2021, 27 March - 1 April, 2021.
White-box fairness testing through adversarial sampling.
ICSE ‘20: 42nd International Conference on Software Engineering, Seoul, South Korea, 27 June - 19 July, 2020.
Global PAC Bounds for Learning Discrete Time Markov Chains.
Computer Aided Verification - 32nd International Conference, CAV 2020, Los Angeles, CA, USA, July 21-24, 2020, Proceedings, Part II.
Adversarial sample detection for deep neural network through model mutation testing.
Proceedings of the 41st International Conference on Software Engineering, ICSE 2019, Montreal, QC, Canada, May 25-31, 2019.
People
SUN Jun
PHAM Hong Long
ZHANG Yueling
SUN Bing
