Parallelism in Deep Learning Accelerators

被引：0

作者：

Song, Linghao ^{[1
]}

Chen, Fan ^{[1
]}

Chen, Yiran ^{[1
]}

Li, Hai Helen ^{[1
]}

机构：

[1] Duke Univ, Durham, NC 27706 USA

来源：

2020 25TH ASIA AND SOUTH PACIFIC DESIGN AUTOMATION CONFERENCE, ASP-DAC 2020 | 2020年

关键词：

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Deep learning is the core of artificial intelligence and it achieves state-of-the-art in a wide range of applications. The intensity of computation and data in deep learning processing poses significant challenges to the conventional computing platforms. Thus, specialized accelerator architectures are proposed for the acceleration of deep learning In this paper, we classify the design space of current deep learning accelerators into three levels, (1) processing engine, (2) memory and (3) accelerator, and present a constructive view from a perspective of parallelism in the three levels.

引用

页码：645 / 650

页数：6

共 50 条

[1] Exploring Parallelism in the Deep Learning Arena
Jamali, Mohsin M.
2018 IEEE 61ST INTERNATIONAL MIDWEST SYMPOSIUM ON CIRCUITS AND SYSTEMS (MWSCAS), 2018, : 170 - 173
[2] Scheduling Dynamic Parallelism on Accelerators
Blagojevic, Filip
Iancu, Costin
Yelick, Katherine
Curtis-Maury, Matthew
Nikolopoulos, Dimitrios S.
Rose, Benjamin
CF'09: CONFERENCE ON COMPUTING FRONTIERS & WORKSHOPS, 2009, : 161 - 170
[3] Ergodic Approximate Deep Learning Accelerators
van Lijssel, Tim
Balatsoukas-Stimming, Alexios
FIFTY-SEVENTH ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS, IEEECONF, 2023, : 734 - 738
[4] Hardware Accelerators for Deep Reinforcement Learning
Mishra, Vinod K.
Basu, Kanad
Arunachalam, Ayush
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING FOR MULTI-DOMAIN OPERATIONS APPLICATIONS V, 2023, 12538
[5] AdequateDL: Approximating Deep Learning Accelerators
Sentieys, Olivier
Filip, Silviu
Briand, David
Novo, David
Dupuis, Etienne
O'Connor, Ian
Bosio, Alberto
2021 24TH INTERNATIONAL SYMPOSIUM ON DESIGN AND DIAGNOSTICS OF ELECTRONIC CIRCUITS & SYSTEMS (DDECS), 2021, : 37 - 40
[6] Model Parallelism optimization with deep reinforcement learning
Mirhoseini, Azalia
2018 IEEE INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW 2018), 2018, : 855 - 855
[7] Exploiting Parallelism Opportunities with Deep Learning Frameworks
Wang, Yu Emma
Wu, Carole-Jean
Wang, Xiaodong
Hazelwood, Kim
Brooks, David
ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION, 2021, 18 (01)
[8] On the Acceleration of Deep Learning Model Parallelism with Staleness
Xu, An
Huo, Zhouyuan
Huang, Heng
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2020, : 2085 - 2094
[9] Chronos: Efficient Speculative Parallelism for Accelerators
Abeydeera, Maleen
Sanchez, Daniel
TWENTY-FIFTH INTERNATIONAL CONFERENCE ON ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS (ASPLOS XXV), 2020, : 1247 - 1262
[10] Exploiting deep learning accelerators for neuromorphic workloads
Sun, Pao-Sheng Vincent
Titterton, Alexander
Gopiani, Anjlee
Santos, Tim
Basu, Arindam
Lu, Wei D.
Eshraghian, Jason K.
NEUROMORPHIC COMPUTING AND ENGINEERING, 2024, 4 (01):

← 1 2 3 4 5 →