A QoS-QoR Aware CNN Accelerator Design Approach

doi:10.1109/TCAD.2018.2877010

	A QoS-QoR Aware CNN Accelerator Design Approach
	Wang, Ying 1,2; Li, Huawei 1,2; Cheng, Long 3; Li, Xiaowei 1,2
	2019-11-01
发表期刊	IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
ISSN	0278-0070
卷号	38 期号:11 页码:1995-2007
摘要	Recently powerful convolutional neural network (CNN) accelerators are emerging as energy-efficient solutions for real-time vision/speech processing, recognition and a wide spectrum of approximate computing applications. In addition to the broad applicability scope of such deep learning (DL) accelerators, we found that the fascinating feature of deterministic performance makes them ideal candidates as application-processors in embedded SoCs concerned with real-time processing. However, unlike traditional accelerator designs, DL accelerators introduce the new aspect of design tradeoff between real-time processing [quality of service (QoS)] and computation approximation [quality of result (QoR)] into embedded systems. This paper proposes an elastic CNN acceleration architecture that automatically adapts to the user-specified QoS constraint by exploiting the error-resilience in typical approximate computing workloads. For the first time, the proposed design, including the network tuning-and-mapping software and reconfigurable accelerator hardware, aims to reconcile the design constraint of QoS and QoR, which are respectively, the critical concerns in real-time and approximate computing. It is shown in experiments the proposed architecture enables the embedded system to work flexibly in an expanded operating space, significantly enhances its real-time ability, and maximizes the system energy-efficiency within the user-specified QoS-QoR constraint through self-reconfiguration. Also, we showcase the application of the proposed design approach to lower power image recognition challenge (LPIRC) and how it is employed to forge an energy-efficient solution to the LPIRC contest.
关键词	Approximate computing convolutional neural network (CNN) deep learning (DL) quality of service (QoS) real-time
DOI	10.1109/TCAD.2018.2877010
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61874124] ; National Natural Science Foundation of China[61504153] ; National Natural Science Foundation of China[61432017] ; National Natural Science Foundation of China[61532017] ; National Natural Science Foundation of China[61402146] ; National Natural Science Foundation of China[61521092] ; YESS Hip Program by CAST
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Interdisciplinary Applications ; Engineering, Electrical & Electronic
WOS记录号	WOS:000505522900001
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/14987
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Li, Huawei; Li, Xiaowei
作者单位	1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 3.Univ Coll Dublin, Dublin D04 V1W8, Ireland
推荐引用方式 GB/T 7714	Wang, Ying,Li, Huawei,Cheng, Long,et al. A QoS-QoR Aware CNN Accelerator Design Approach[J]. IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,2019,38(11):1995-2007.
APA	Wang, Ying,Li, Huawei,Cheng, Long,&Li, Xiaowei.(2019).A QoS-QoR Aware CNN Accelerator Design Approach.IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,38(11),1995-2007.
MLA	Wang, Ying,et al."A QoS-QoR Aware CNN Accelerator Design Approach".IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS 38.11(2019):1995-2007.