Jiawen Liu


PhD Candidate, EECS

University of California, Merced

Email: jliu265@ucmerced.edu


Biography

I'm a PhD candidate at UC Merced supervised by Prof. Dong Li. My research interests mainly focus on computer system, high performance computing and machine learning with a target on building high-performance and resource-aware systems for machine learning and tensor computation.


Recent News

07/2020: Successfully pass qualifying exam.
06/2020: Receive Graduate Travel Award from UC Merced.
05/2020: Start summer research intern at Pacific Northwest National Laboratory (PNNL).
03/2020: One paper, "RIANN: Real-time Incremental Learning with Approximate Nearest Neighbor on Mobile Devices", is accepted in USENIX OpML'20.
02/2020: One paper, "Flame: A Self-Adaptive Auto-Labeling System for Heterogeneous Mobile Processors", is accepted in MLSys-W'20.
09/2019: One paper, "Performance Analysis and Characterization of Training Deep Learning Models on Mobile Device", is accepted in ICPADS'19.
03/2019: Receive IEEE Travel Award from Technical Committee on Parallel Processing (TCPP) for IPDPS'19.
01/2019: One paper, "Runtime Concurrency Control and Operation Scheduling for High Performance Neural Network Training",is accepted in IPDPS'19.
10/2018: Receive ACM/IEEE Travel Award from Advanced Computing Systems Association (USENIX) for OSDI'18.
09/2018: Receive ACM/IEEE Travel Award from National Science Foundation (NSF) for MICRO'18.
08/2018: One paper, "Processing in Memory for Energy-efficient CNN Training: A Heterogeneous Approach", is accepted in MICRO'18.
More >

Publications


RIANN: Real-time Incremental Learning with Approximate Nearest Neighbor on Mobile Devices
Jiawen Liu, Zhen Xie, Dimitrios Nikolopoulos, Dong Li
USENIX Conference on Operational Machine Learning (USENIX OpML), 2020.


Flame: A Self-Adaptive Auto-Labeling System for Heterogeneous Mobile Processors
Jiawen Liu, Jie Liu, Zhen Xie, Dong Li
On-Device Intelligence Workshop at Machine Learning and Systems Conference (MLSys-W), 2020.


Runtime Concurrency Control and Operation Scheduling for High Performance Neural Network Training
Jiawen Liu, Dong Li, Gokcen Kestor, Jeffrey Vetter
IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS), 2019. (Acceptance rate: 103/372=27%)


Performance Analysis and Characterization of Training Deep Learning Models on Mobile Device
Jie Liu, Jiawen Liu, Wan Du, Dong Li
IEEE 25th International Conference on Parallel and Distributed Systems (ICPADS), 2019. (Acceptance rate: 97/346=28%)


Processing in Memory for Energy-efficient CNN Training: A Heterogeneous Approach
Jiawen Liu, Hengyu Zhao, Matheus A. Ogleari, Dong Li, Jishen Zhao
IEEE/ACM 51st International Symposium on Microarchitecture (MICRO), 2018. (Acceptance rate: 74/348=21%)


HOMP: Automated Distribution of Parallel Loops and Data in Highly Parallel Accelerator-Based Systems
Yonghong Yan, Jiawen Liu, Kirk Cameron, Mariam Umar
IEEE 31st International Parallel and Distributed Processing Symposium (IPDPS), 2017. (Acceptance rate: 116/516=22%)


Teaching

Spring 2020: CSE021 - Introduction to Computing II
Spring 2019: CSE179 - Introduction to Parallel Computing

Professional Services

External reviewers: IPDPS'20, TACO’19, ICPP’19, Cluster’19, NPC’19, SC’18, CCGrid’17, ISCHPC’16, PMAM’16, etc

Work Experience

Summer 2020: Research intern at Pacific Northwest National Laboratory (PNNL)
Summer 2018: Research intern at Oak Ridge National Laboratory (ORNL)


Last updated on 07/2020.