Yuan Feng  

I'm a 3-year Ph.D student in University of California, Merced, where I am so fortunately working with Prof. Hyeran Jeon. Prior to that, I received my M.Sc from University of Edinburgh, and my B.E. from Xiamen University (those are all pretty beautifull cities, be sure to visit them!!).

My research areas are Computer Architecture (more specifically, GPGPU Architecture, Memory systems and their interaction with ML). I design high performance, robust, and engergy-efficient Hardware Architecture and Systems. My recent work focuses on designing architecture support for Multi-Chip GPU

Contact: yfeng44 [at] ucmerced [dot] edu

Resume    /    LinkedIn

March 2024 One paper got accepted by SC'2024 !
March 2024 One paper got accepted by ISCA'2024 ! Stay tuned!
June 2023 Present our collaboration work AttnMemo at Western Digital Research.
Publications and Presentations
[SC'24] Efficient Tensor Offloading for Large Deep-Learning Model Training based on Compute Express Link
Dong Xu, Yuan Feng, Kwangsik Shin, Daewoo Kim, Hyeran Jeon and Dong Li
[ISCA'24] Barre Chord: Efficient Virtual Memory Translation for Multi-Chip-Module GPUs
Yuan Feng, Seonjin Na, Hyesoon Kim, and Hyeran Jeon
[IISWC'23] Regular-Decepticon: Understanding Vulnerabilities of Transformers
Mujahid Al Rafi, Yuan Feng, Fan Yao, Meng Tang, Hyeran Jeon
[arxiv Preprint] MEMO : Accelerating Transformers with Memoization on Big Memory Systems
Yuan Feng, Hyeran Jeon, Filip Blagojevic, Cyril Guyot, Qing Li, and Dong Li
[GPGPU'23] Understanding Scalability of Multi-GPU Systems
Yuan Feng , Hyeran Jeon
[MICRO'22 SRC] Accelerate Transformers with Efficient Memoization
Yuan Feng , Hyeran Jeon, Dong Li
[NOPE'22, ASPLOS Workshop] Too Noisy To Extract: Pitfalls of Model Extraction Attacks
Mujahid Al Rafi, Yuan Feng , Hyeran Jeon
Summer, 2024 Research Intern, Futurewei (Mentor: Sang Wook Do)
Summer, 2023 Research Intern, Alibaba Cloud (Mentor: Jian Chen and Qixiao Liu)
Professional Activities
GPGPU'23,'24 Web Chair
Conference Travel Grant MICRO' 22, ISCA' 24
UC Merced Bobcat Graduate Research Fellowship 2022,2023

credit Kai Wu, Jon Barron & Limin Yang