Artificial Intelligence & Risk Laboratory

Led by Prof Jize Zhang, the Artificial Intelligence & Risk Laboratory at Hong Kong University of Science and Technology leverages cutting-edge AI and computational science to tackle pressing real-world problems.

Our work spans developing reliable, efficient AI foundation model-driven methods for robust infrastructure defect detection from drone imagery, to creating scalable, GPU-accelerated neural models for rapid and accurate natural hazard modeling.

Join us: We always welcome applications from talented researchers. Please refer to Apply.

Our work

Our news

01-Apr-2025: Prof Zhang invited to the Editorial Board of Data-Centric Engineering   (more)

29-Jun-2024: RGC Early Career Scheme awarded to Professor Zhang.   (more)

02-Jun-2024: Prof Zhang participated in SHRCC2024 at Hawaii.   (more)

27-Apr-2024: PhD student Xi Zhong won Best Student Paper Award in ICVRAM-ISUMA 2024.   (more)

10-Apr-2024: Prof Zhang invited to present for the 2024 Engineering Structures (Asia-Pacific) Young Scientist Forum.   (more)

Our priorities

Low Altitude Economy
Low Altitude Economy

Control, sensing, and decision-making for UAVs.

Uncertainty Quantification
Uncertainty Quantification

UQ for large-scale AI models and engineering solvers.

Natural Hazard Modeling
Natural Hazard Modeling

Scalable and efficient models for natural hazards.

Our selected publications

SelectSeg: Uncertainty-based selective training and prediction for accurate crack segmentation under limited data and noisy annotations  (2025)  ·  Chen Zhang, Mahdi Bahrami, …, Yantao Yu, Jize Zhang  ·  Reliability Engineering & System Safety
TC-SINDy: Improving physics-based deterministic tropical cyclone track and intensity model via data-driven sparse identification of Nonlinear Dynamics  (2024)  ·  Xi Zhong, Wenjun Jiang, Jize Zhang  ·  Journal of Wind Engineering and Industrial Aerodynamics
Mix-n-Match: Ensemble and Compositional Methods for Uncertainty Calibration in Deep Learning  (2020)  ·  Jize Zhang, Bhavya Kailkhura, T. Yong-Jin Han  ·  International Conference on Machine Learning (ICML)

Our funders