معرفی
Yanghao Zhang is a Researcher at Imperial College London's Department of Computing, part of the Faculty of Engineering, working under Prof. Alessio Lomuscio at the Safe AI Lab (SAIL). He also holds a Research Scientist position at Safe Intelligence. He completed his PhD at the University of Liverpool (2020–2024), supervised by Prof. Wenjie Ruan and Prof. Xiaowei Huang, with earlier PhD studies at the University of Exeter. His research focuses on adversarial machine learning, safety verification, robust deep learning applications, and graph neural networks.
- Education: MSc Artificial Intelligence (Distinction, University of Southampton); BEng Software Engineering (Huizhou University).
His work emphasizes adversarial attacks/defenses, semantic robustness in perception systems, and safety of large language models. Key contributions include frameworks for evaluating robustness in autonomous systems and risk-aware learning algorithms. Recent achievements include AAAI and CVPR publications, invitations as a reviewer for top conferences (e.g., AISTATS, ICML, NeurIPS), and PhD defense in 2024. He collaborates with industry (Safe Intelligence) and academia to advance trustworthy AI systems.
Labs/Teams: Primary affiliation with Safe AI Lab (SAIL) at Imperial, and research at Safe Intelligence. Active in reviewing for journals like TKDE and conferences like ICLR, CVPR, and NeurIPS.
