Guangdong Liu
Papers
1
Total Citations
21
H-Index
1
About
Guangdong Liu is a leading researcher in high-performance computing and FPGA-based acceleration, with a primary focus on convolutional neural network (CNN) accelerators. His most impactful work, "XVDPU: A High Performance CNN Accelerator on the Versal Platform Powered by the AI Engine" (2022), has garnered 21 citations and represents a breakthrough in addressing the critical bottlenecks of computation and I/O bandwidth in modern deep learning. By leveraging Xilinx's cutting-edge 7nm Versal platform and its AI Engine architecture, Liu's design achieves unprecedented efficiency for large-scale CNN inference, directly tackling the challenges posed by increasingly complex networks and higher-resolution inputs. This work is notable for its practical implementation on state-of-the-art hardware, offering a scalable solution for real-world computer vision applications. Liu's contributions are pivotal for researchers and engineers seeking to push the boundaries of real-time AI processing, demonstrating how novel hardware-software co-design can overcome the limitations of traditional GPU-based systems. His research continues to influence the next generation of domain-specific accelerators, making him a key figure in the evolution of efficient deep learning hardware.
Research Focus
Key Achievements
Top Papers
- 1