Huan Zhang 2022 Early Career Fellow
Hard Problem Develop more capable and more general AI, that is useful, safe and earns public trust

Huan Zhang is an Assistant Professor at the University of Illinois Urbana-Champaign. He received his Ph.D. degree at UCLA in 2020. Huan’s research focuses on the trustworthiness of AI, especially on developing formal verification methods for deep neural networks. Huan led a multi-institutional team that won the 2021 and 2022 International Verification of Neural Networks Competition. Huan was awarded an IBM Ph.D. fellowship in 2018, and received the 2021 Adversarial Machine Learning (AdvML) Rising Star Award.

AI2050 Project

The use of artificial intelligence (AI) is challenging in mission-critical scenarios such as aircraft control, autonomous driving, and medical procedures because AI is often an untrustable black-box. Through this fellowship, Huan’s research aims to improve and guarantee the trustworthiness of AI, making AI safer, more robust, more predictable and more reliable. Huan will focus on developing formal verification methods that can rigorously prove the trustworthiness of AI, enabling the use of AI in mission-critical systems with guaranteed performance.

Project artifacts

L. Yang, H. Dai, Z. Shi, C. Hsieh, R. Tedrake, and H. Zhang. Lyapunov-stable neural control for state and output feedback: a novel formulation for efficient synthesis and verification. arXiv. 2024.

AI2050 Community Perspective — Huan Zhang (2024)

X. Guo, F. Yu, H. Zhang, L. Qin, and B. Hu. COLD-Attack: jailbreaking LLMs with stealthiness and controllability. arXiv. 2024.

S. Kotha, C. Brix, Z. Kolter, K. Dvijotham, and H. Zhang. Provably bounding neural network preimages. NeurIPS. 2023.