Yan Wang

About Me

I’m a PhD student at Institute for Computing System Architecture (ICSA), School of Informatics, University of Edinburgh, working under the mentorship of Jingjie Li and Paul Patras . Before that, I was a master student at State Key Laboratory of Cyberspace Security Defense, Institute of Information Engineering, Chinese Academy of Sciences (CAS), working under the mentorship of Prof. Rui Hou . I obtained my Bachelor’s degree with major in Software Engineering from the School of Software at Shandong University in 2022.

My research lies at the intersection of computer architecture, machine learning, and security, with a focus on leveraging advanced computing architectures to develop machine learning applications for security-critical scenarios.

Research Interests

Machine Learning Systems: Exploring the design and optimization of efficient machine learning inference/training systems, particularly for resource-constrained environments and real-time applications.
Adversarial Machine Learning: Investigating strategies to mitigate the vulnerabilities of machine learning models against adversarial attacks, ensuring their robustness and reliability in security-sensitive applications.
LLM Security: Delving into the security challenges posed by large language models (LLMs), developing techniques to safeguard these powerful AI models from misuse and malicious exploitation.

News

[Oct. 2025] Yan admitted to the PhD program at the School of Informatics, University of Edinburgh, fully funded by the EPSRC DLA Scholarship!
[Dec. 2024] Two papers co-authored by Yan accepted to ICASSP 2025!
[Nov. 2024] Yan received the Dean’s Special Scholarship of the Institute of Information Engineering, Chinese Academy of Sciences (top 0.5%), the highest honor at the institute!
[Nov. 2024] Yan received the National Scholarship for Graduate Students in China (top 0.2%)!
[Oct. 2024] Our work ROBIN: A Novel Framework for Accelerating Robust Multi-Variant Training accepted to ASP-DAC 2025! See you in Tokyo!
[May 2024] Our work FakeGuard: A Novel Accelerator Architecture for Deepfake Detection Networks accepted to Euro-Par 2024!
[Apr. 2024] Yan was selected to join DAC Young Fellow program at DAC 2024!

Selected Publications

IEEE TCAD

EnsGuard: A Novel Acceleration Framework for Adversarial Ensemble Learning

Xingbin Wang*, Yan Wang*(equal contribution), Yulan Su, Sisi Zhang, Dan Meng, Rui Hou

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems(TCAD), April. 2024.
DAC

Garrison: A High-Performance GPU-Accelerated Inference System for Adversarial Ensemble Defense

Yan Wang, Xingbin Wang, Zechao Lin, Yulan Su, Sisi Zhang, Rui Hou, Dan Meng

IEEE/ACM Design Automatic Conference(DAC), June. 2024.
ASP-DAC

Robin: A Novel Framework for Accelerating Robust Multi-Variant Training

Yan Wang, Xingbin Wang, Yulan Su, Sisi Zhang, Zechao Lin, Dan Meng, Rui Hou

Asia and South Pacific Design Automation Conference (ASP-DAC), January. 2025.

Fun Fact

History loops in curious ways. My academic great-grand-advisor, Dr. Xia Peisu , earned her PhD at the University of Edinburgh in 1950. Now, seventy-five years later, I find myself in the same city.