TY - JOUR
T1 - Review of Optimal Convolutional Neural Network Accelerator Platforms for Mobile Devices
AU - Kim, Hyun
N1 - Publisher Copyright:
© 2022. The Korean Institute of Information Scientists and Engineers
PY - 2022/6
Y1 - 2022/6
N2 - In recent years, convolutional neural networks (CNNs) have achieved remarkable performance enhancement, and researchers have endeavored to use CNN applications on power-constrained mobile devices. Accordingly, low-power and high-performance CNN accelerators for mobile devices are receiving significant attention. This paper presents the overall process of designing optimal CNN accelerator platforms for mobile devices based on algorithm, architecture, and memory system co-design while introducing various existing studies related to specific research fields.
AB - In recent years, convolutional neural networks (CNNs) have achieved remarkable performance enhancement, and researchers have endeavored to use CNN applications on power-constrained mobile devices. Accordingly, low-power and high-performance CNN accelerators for mobile devices are receiving significant attention. This paper presents the overall process of designing optimal CNN accelerator platforms for mobile devices based on algorithm, architecture, and memory system co-design while introducing various existing studies related to specific research fields.
KW - Convolutional neural networks
KW - Hardware accelerator
KW - Lowpower memory
KW - Mobile device
KW - Network compression
UR - https://www.scopus.com/pages/publications/85133765622
U2 - 10.5626/JCSE.2022.16.2.113
DO - 10.5626/JCSE.2022.16.2.113
M3 - Article
AN - SCOPUS:85133765622
SN - 1976-4677
VL - 16
SP - 113
EP - 119
JO - Journal of Computing Science and Engineering
JF - Journal of Computing Science and Engineering
IS - 2
ER -