Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Xinheng Liu

Xinheng Liu

HiKonv: Maximizing the Throughput of Quantized Convolution With Novel Bit-wise Management and Computation

Jul 22, 2022
Yao Chen, Junhao Pan, Xinheng Liu, Jinjun Xiong, Deming Chen

Figure 1 for HiKonv: Maximizing the Throughput of Quantized Convolution With Novel Bit-wise Management and Computation

Figure 2 for HiKonv: Maximizing the Throughput of Quantized Convolution With Novel Bit-wise Management and Computation

Figure 3 for HiKonv: Maximizing the Throughput of Quantized Convolution With Novel Bit-wise Management and Computation

Figure 4 for HiKonv: Maximizing the Throughput of Quantized Convolution With Novel Bit-wise Management and Computation

* The conference version is pubilished in Proceedings of ASP-DAC 2022. arXiv admin note: substantial text overlap with arXiv:2112.13972

Access Paper or Ask Questions

HiKonv: High Throughput Quantized Convolution With Novel Bit-wise Management and Computation

Dec 28, 2021
Xinheng Liu, Yao Chen, Prakhar Ganesh, Junhao Pan, Jinjun Xiong, Deming Chen

Figure 1 for HiKonv: High Throughput Quantized Convolution With Novel Bit-wise Management and Computation

Figure 2 for HiKonv: High Throughput Quantized Convolution With Novel Bit-wise Management and Computation

Figure 3 for HiKonv: High Throughput Quantized Convolution With Novel Bit-wise Management and Computation

Figure 4 for HiKonv: High Throughput Quantized Convolution With Novel Bit-wise Management and Computation

* 7 pages, 6 figures. Accepted by ASP-DAC 2022

Access Paper or Ask Questions

WinoCNN: Kernel Sharing Winograd Systolic Array for Efficient Convolutional Neural Network Acceleration on FPGAs

Jul 09, 2021
Xinheng Liu, Yao Chen, Cong Hao, Ashutosh Dhar, Deming Chen

Figure 1 for WinoCNN: Kernel Sharing Winograd Systolic Array for Efficient Convolutional Neural Network Acceleration on FPGAs

Figure 2 for WinoCNN: Kernel Sharing Winograd Systolic Array for Efficient Convolutional Neural Network Acceleration on FPGAs

Figure 3 for WinoCNN: Kernel Sharing Winograd Systolic Array for Efficient Convolutional Neural Network Acceleration on FPGAs

Figure 4 for WinoCNN: Kernel Sharing Winograd Systolic Array for Efficient Convolutional Neural Network Acceleration on FPGAs

* Published in the proceedings of ASAP 2021

Access Paper or Ask Questions

FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activations

Dec 22, 2020
Yichi Zhang, Junhao Pan, Xinheng Liu, Hongzheng Chen, Deming Chen, Zhiru Zhang

Figure 1 for FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activations

Figure 2 for FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activations

Figure 3 for FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activations

Figure 4 for FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activations

* Published at the 29th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA 2021)

Access Paper or Ask Questions

EDD: Efficient Differentiable DNN Architecture and Implementation Co-search for Embedded AI Solutions

May 06, 2020
Yuhong Li, Cong Hao, Xiaofan Zhang, Xinheng Liu, Yao Chen, Jinjun Xiong, Wen-mei Hwu, Deming Chen

Figure 1 for EDD: Efficient Differentiable DNN Architecture and Implementation Co-search for Embedded AI Solutions

Figure 2 for EDD: Efficient Differentiable DNN Architecture and Implementation Co-search for Embedded AI Solutions

Figure 3 for EDD: Efficient Differentiable DNN Architecture and Implementation Co-search for Embedded AI Solutions

Figure 4 for EDD: Efficient Differentiable DNN Architecture and Implementation Co-search for Embedded AI Solutions

* Accepted by Design Automation Conference (DAC'2020)

Access Paper or Ask Questions

Face Recognition with Hybrid Efficient Convolution Algorithms on FPGAs

Mar 23, 2018
Chuanhao Zhuge, Xinheng Liu, Xiaofan Zhang, Sudeep Gummadi, Jinjun Xiong, Deming Chen

Figure 1 for Face Recognition with Hybrid Efficient Convolution Algorithms on FPGAs

Figure 2 for Face Recognition with Hybrid Efficient Convolution Algorithms on FPGAs

Figure 3 for Face Recognition with Hybrid Efficient Convolution Algorithms on FPGAs

Figure 4 for Face Recognition with Hybrid Efficient Convolution Algorithms on FPGAs

* This paper is accepted in GLSVLSI'18

Access Paper or Ask Questions