A Review of FPGA-Based Custom Computing Architecture for Convolutional Neural Network Inference
A Review of FPGA-Based Custom Computing Architecture for Convolutional Neural Network Inference作者机构：School of Electronics and Information Engineering Harbin Institute of Technology
出 版 物：《Chinese Journal of Electronics》 (电子学报(英文))
年 卷 期：2021年第30卷第1期
页 面：1-17页
核心收录：
学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 080902[工学-电路与系统] 0809[工学-电子科学与技术（可授工学、理学学位）] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）]
基 金：supported by the National Natural Science Foundation of China (No.61803121) the Postdoctoral Science Foundation of China (No.2019M651277)
主 题：inference mechanisms computing resources hardware-oriented model compression CNN structure design computing overhead convolutional neural nets electronic engineering computing field-programmable gate array-based custom computing architecture computation complexity FPGA-based custom computing architecture design computational complexity convolutional neural network inference CNN inference performance power consumption field programmable gate arrays
摘 要：Convolutional neural network(CNN)has been widely adopted in many tasks. Its inference process is usually applied on edge devices where the computing resources and power consumption are *** present, the performance of general processors cannot meet the requirement for CNN models with high computation complexity and large number of parameters. Field-programmable gate array(FPGA)-based custom computing architecture is a promising solution to further enhance the CNN inference *** software/hardware co-design can effectively reduce the computing overhead, and improve the inference performance while ensuring accuracy. In this paper, the mainstream methods of CNN structure design, hardwareoriented model compression and FPGA-based custom architecture design are summarized, and the improvement of CNN inference performance is demonstrated through an example. Challenges and possible research directions in the future are concluded to foster research efforts in this domain.
同方期刊数据库