Implementing Binarized Neural Network Processor on FPGA-Based Platform

被引：0

作者：

Lee, Jeahack ^{[1
]}

Kim, Hyeonseong ^{[1
]}

Kim, Byung-Soo ^{[1
]}

Jeon, Seokhun ^{[1
]}

Lee, Jung Chul ^{[1
]}

Kim, Dong Sun ^{[1
]}

机构：

[1] Korea Elect Technol Inst, SoC Platform Res Ctr, Seongnam Si, South Korea

来源：

2022 IEEE INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE CIRCUITS AND SYSTEMS (AICAS 2022): INTELLIGENT TECHNOLOGY IN THE POST-PANDEMIC ERA | 2022年

关键词：

Deep Learning; Binarized Neural Network; Neural Network Processor; FPGA Accelerator;

D O I：

10.1109/AICAS54282.2022.9869997

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Binarized neural networks (BNNs) have 1-bit weights and activations, which are well suited for FPGAs. The BNNs suffer from accuracy loss compared with conventional neural networks. Shortcut connections are introduced to address the performance degradation. This work proposes a BNN processor supporting the shortcut connects. To evaluate the performance of the processor, we implement the system on an FPGA (Xilinx Kintex UltraScale). Our experiments show that the proposed processor achieves state-of-the-art energy efficiency.

引用

页码：469 / 471

页数：3

共 50 条

[1] A High-Efficiency FPGA-Based Accelerator for Binarized Neural Network
Guo, Peng
Ma, Hong
Chen, Ruizhi
Wang, Donglin
[J]. JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS, 2019, 28
[2] Neural network processor for a FPGA-based multiband fluorometer device
Tabari, Karima
Boukadoum, Mounir
Bensaoula, Abdelhak
Starikov, David
[J]. 2006 INTERNATIONAL WORKSHOP ON COMPUTER ARCHITECTURE FOR MACHINE PERCEPTION AND SENSING, 2006, : 198 - +
[3] An FPGA-based Accelerator Platform Implements for Convolutional Neural Network
Meng, Xiao
Yu, Lixin
Qin, Zhiyong
[J]. 2019 THE 3RD INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPILATION, COMPUTING AND COMMUNICATIONS (HP3C 2019), 2019, : 25 - 28
[4] Customizable FPGA-based Accelerator for Binarized Graph Neural Networks
Wang, Ziwei
Que, Zhiqiang
Luk, Wayne
Fan, Hongxiang
[J]. 2022 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS 22), 2022, : 1968 - 1972
[5] Optical processor for a binarized neural network
Huang, Long
Yao, Jianping
[J]. OPTICS LETTERS, 2022, 47 (15) : 3892 - 3895
[6] An FPGA-Based Processor for Training Convolutional Neural Networks
Liu, Zhiqiang
Dou, Yong
Jiang, Jingfei
Wang, Qiang
Chow, Paul
[J]. 2017 INTERNATIONAL CONFERENCE ON FIELD PROGRAMMABLE TECHNOLOGY (ICFPT), 2017, : 207 - 210
[7] An FPGA-based Application-Specific Processor for Implementing the Exponential Function
Jadhav, Shrikant S.
Gloster, Clay
Naher, Jannatun
Doss, Christopher
Kim, Youngsoo
[J]. IEEE SOUTHEASTCON 2020, 2020,
[8] BNNsplit: Binarized Neural Networks for embedded distributed FPGA-based computing systems
Fiscaletti, Giorgia
Speziali, Marco
Stornaiuolo, Luca
Santambrogio, Marco D.
Sciuto, Donatella
[J]. PROCEEDINGS OF THE 2020 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE 2020), 2020, : 975 - 978
[9] FPGA-based SIMD processor
Li, SYC
Cheuk, GCK
Lee, KH
Leong, PHW
[J]. FCCM 2003: 11TH ANNUAL IEEE SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES, PROCEEDINGS, 2003, : 267 - 268
[10] OPU: An FPGA-Based Overlay Processor for Convolutional Neural Networks
Yu, Yunxuan
Wu, Chen
Zhao, Tiandong
Wang, Kun
He, Lei
[J]. IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 2020, 28 (01) : 35 - 47

← 1 2 3 4 5 →