Fanghui Liu

Dr. Fanghui Liu is an Assistant Professor at the Department of Computer Science, University of Warwick, UK, and a faculty member of the Centre for Discrete Mathematics and its Applications (DIMAP). Previously, he was a postdoctoral researcher at EPFL (2021-2023) and KU Leuven (2019-2021). He holds a B.E. in Automation from Harbin Institute of Technology (2014) and a Ph.D. in Machine Learning from Shanghai Jiao Tong University (2019).

His research focuses on the theoretical foundations of modern machine learning, including kernel methods, adversarial robustness, neural architecture search, and large language models. Notable contributions include work on adversarial training, robust NAS benchmarks, and theoretical analysis of fine-tuning techniques.

He has received grants from the Royal Society, Alan Turing Institute, and DAAD. Awards include the DAAD AInet Fellowship (2024) and recognition as a Notable Reviewer for ICLR 2023. His research has been published in top conferences like ICML, ICLR, and NeurIPS, with a focus on bridging theory and practice in machine learning.