Bingyi Zhang
University of Southern California, FPGA/PARALLEL COMPUTING LAB
EEB 244
3740 McClintock Ave
Los Angeles, CA 90089
I am a fifth-year Ph.D. candidate in the Ming Hsieh Department of Electrical Engineering at USC. Currently, I am under the advisement of professor Viktor K. Prasanna. My research interest is high performance computing (HPC), computer architecture, reconfigurable computing, graph machine learning and VLSI. Currently, I focus on sparse linear algebra acceleration on FPGA and graph neural network acceleration on FPGA.
Before I joined USC, I persued my master degree at State Key Laboratory of ASIC and System at Fudan Universty, under the guidance of Professor Jun Han and Professor Xiaoyang Zeng.
*I am glad to serve as a reviewer for journal or conference papers, as well as a TPC member for conference or workshop, in the area of High Performance Computing, Reconfigurable Computing, Circuit and System, Digital Signal Processing, VLSI, Deep Learning, Machine Learning, Graph Analytics and Computing, etc. Please feel free to contact me through email: bingyizh@usc.edu
I have been invited and served as the reviewer for more than (>) 80 times for various journals and conferences, including:
-
Journal (> 70 times): IEEE Transactions on Very Large Scale Integration Systems (TVLSI); IEEE Transactions on Image Processing (TIP); IET Computer Vision; IET Image Processing; Neurocomputing (NEUCOM); Engineering Applications of Artificial Intelligence (EAAI); Information Sciences; Journal of Electronic Science and Technology; Microelectronics Journal; Microprocessors and Microsystems;
-
Conference (>10 papers): ASICON 2021; ICSICT 2022; IEEE international radar conference 2023;
news
Jun 16, 2023 | Our paper “GraphAGILE: An FPGA-based Overlay Accelerator for Low-latency GNN Inference” is accepted by IEEE Transactions on Parallel and Distributed Systems |
---|---|
Jun 4, 2023 | Our paper “Exploiting On-chip Heterogeneity of Versal Architecture for GNN Inference Acceleration” is accepted by 33nd International Conference on Field Programmable Logic and Applications (FPL 2023) |
May 20, 2023 | Excited to share that I have been awarded 1st place for the Outstanding Poster Award at the IPDPS 2023 PhD Forum! |
May 8, 2023 | I am organizing the FCCM 2023 conference (https://www.fccm.org/) as the Local Arrangements Chair. The conference has finished successfully! |
Jan 27, 2023 | My paper “Dynasparse: Accelerating GNN Inference through Dynamic Sparsity Exploitation” is accepted by 37th IEEE International Parallel and Distributed Processing Symposium (IPDPS 2023) |
Sep 16, 2022 | My paper “Low-latency Mini-batch GNN Inference on CPU-FPGA Heterogeneous Platform” is accepted by 2022 International Conference on High Performance Computing, Data, and Analytics (HiPC 2022) |
Aug 16, 2022 | My paper “Performance Modeling Sparse MTTKRP Using Optical Static Random Access Memory on FPGA” is accepted by 26th Annual IEEE High Performance Extreme Computing Virtual Conference (HPEC 2022) |
Jun 14, 2022 | My paper “Accurate, Low-latency, Efficient SAR Automatic Target Recognition on FPGA” is accepted by 32nd International Conference on Field Programmable Logic and Applications (FPL 2022) |
selected publications
- ASAPHardware acceleration of large scale gcn inferenceIn 2020 IEEE 31st International Conference on Application-specific Systems, Architectures and Processors (ASAP), 2020
- FCCMBoostGCN: A framework for optimizing GCN inference on FPGAIn 2021 IEEE 29th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM), 2021
- FPGAHP-GNN: Generating High Throughput GNN Training Implementation on CPU-FPGA Heterogeneous PlatformIn Proceedings of the 2022 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays, 2022
- IPDPSModel-Architecture Co-Design for High Performance Temporal GNN Inference on FPGAIn 2022 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2022
- IPDPSDynasparse: Accelerating GNN Inference through Dynamic Sparsity Exploitation2023 International Parallel and Distributed Processing Symposium, 2023
- TPDSGraphAGILE: An FPGA-based Overlay Accelerator for Low-latency GNN InferenceIEEE Transactions on Parallel and Distributed Systems, 2023