A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems

Jaehyeong Sim; Jun Seok Park; Minhye Kim; Dongmyung Bae; Yeongjae Choi; Lee Sup Kim

doi:10.1109/ISSCC.2016.7418008

A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems

Jaehyeong Sim, Jun Seok Park, Minhye Kim, Dongmyung Bae, Yeongjae Choi, Lee Sup Kim

Computer Science & Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

119 Scopus citations

Abstract

Transmitting massive amounts of image and audio data acquired by Internet-of-Everything (IoE) devices to data center servers for intelligent recognition processes is impractical for energy reasons, requiring in-situ processing of such data. However, algorithms accelerated by previous recognition processors [1, 2] are limited to specific applications, therefore, each IoE device may require an application-specific accelerator. On the other hand, deep convolutional neural networks (CNNs) [3] are a promising machine-learning approach, showing state-of-the-art recognition accuracy in a wide variety of applications, including both image and audio recognition. This makes CNNs a suitable candidate for a universal recognition platform for IoE devices, as described in Fig. 14.6.1. Due to the computational complexity and significant memory requirements of CNNs, a microcontroller unit (MCU) typically used for IoE devices is incapable of producing a meaningful recognition result in an energy-efficient way. Hence, the implementation of an energy-efficient CNN processor is desired to realize intelligent IoE systems.

Original language	English
Title of host publication	2016 IEEE International Solid-State Circuits Conference, ISSCC 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	264-265
Number of pages	2
ISBN (Electronic)	9781467394666
DOIs	https://doi.org/10.1109/ISSCC.2016.7418008
State	Published - 23 Feb 2016
Event	63rd IEEE International Solid-State Circuits Conference, ISSCC 2016 - San Francisco, United States Duration: 31 Jan 2016 → 4 Feb 2016

Publication series

Name	Digest of Technical Papers - IEEE International Solid-State Circuits Conference
Volume	59
ISSN (Print)	0193-6530

Conference

Conference	63rd IEEE International Solid-State Circuits Conference, ISSCC 2016
Country/Territory	United States
City	San Francisco
Period	31/01/16 → 4/02/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Access to Document

10.1109/ISSCC.2016.7418008

Cite this

Sim, J., Park, J. S., Kim, M., Bae, D., Choi, Y., & Kim, L. S. (2016). A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems. In 2016 IEEE International Solid-State Circuits Conference, ISSCC 2016 (pp. 264-265). Article 7418008 (Digest of Technical Papers - IEEE International Solid-State Circuits Conference; Vol. 59). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISSCC.2016.7418008

Sim, Jaehyeong ; Park, Jun Seok ; Kim, Minhye et al. / A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems. 2016 IEEE International Solid-State Circuits Conference, ISSCC 2016. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 264-265 (Digest of Technical Papers - IEEE International Solid-State Circuits Conference).

@inproceedings{30d5667bb44b4c03b56dab194193e6c6,

title = "A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems",

abstract = "Transmitting massive amounts of image and audio data acquired by Internet-of-Everything (IoE) devices to data center servers for intelligent recognition processes is impractical for energy reasons, requiring in-situ processing of such data. However, algorithms accelerated by previous recognition processors [1, 2] are limited to specific applications, therefore, each IoE device may require an application-specific accelerator. On the other hand, deep convolutional neural networks (CNNs) [3] are a promising machine-learning approach, showing state-of-the-art recognition accuracy in a wide variety of applications, including both image and audio recognition. This makes CNNs a suitable candidate for a universal recognition platform for IoE devices, as described in Fig. 14.6.1. Due to the computational complexity and significant memory requirements of CNNs, a microcontroller unit (MCU) typically used for IoE devices is incapable of producing a meaningful recognition result in an energy-efficient way. Hence, the implementation of an energy-efficient CNN processor is desired to realize intelligent IoE systems.",

author = "Jaehyeong Sim and Park, {Jun Seok} and Minhye Kim and Dongmyung Bae and Yeongjae Choi and Kim, {Lee Sup}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 63rd IEEE International Solid-State Circuits Conference, ISSCC 2016 ; Conference date: 31-01-2016 Through 04-02-2016",

year = "2016",

month = feb,

day = "23",

doi = "10.1109/ISSCC.2016.7418008",

language = "English",

series = "Digest of Technical Papers - IEEE International Solid-State Circuits Conference",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "264--265",

booktitle = "2016 IEEE International Solid-State Circuits Conference, ISSCC 2016",

}

Sim, J, Park, JS, Kim, M, Bae, D, Choi, Y & Kim, LS 2016, A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems. in 2016 IEEE International Solid-State Circuits Conference, ISSCC 2016., 7418008, Digest of Technical Papers - IEEE International Solid-State Circuits Conference, vol. 59, Institute of Electrical and Electronics Engineers Inc., pp. 264-265, 63rd IEEE International Solid-State Circuits Conference, ISSCC 2016, San Francisco, United States, 31/01/16. https://doi.org/10.1109/ISSCC.2016.7418008

A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems. / Sim, Jaehyeong; Park, Jun Seok; Kim, Minhye et al.
2016 IEEE International Solid-State Circuits Conference, ISSCC 2016. Institute of Electrical and Electronics Engineers Inc., 2016. p. 264-265 7418008 (Digest of Technical Papers - IEEE International Solid-State Circuits Conference; Vol. 59).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems

AU - Sim, Jaehyeong

AU - Park, Jun Seok

AU - Kim, Minhye

AU - Bae, Dongmyung

AU - Choi, Yeongjae

AU - Kim, Lee Sup

PY - 2016/2/23

Y1 - 2016/2/23

N2 - Transmitting massive amounts of image and audio data acquired by Internet-of-Everything (IoE) devices to data center servers for intelligent recognition processes is impractical for energy reasons, requiring in-situ processing of such data. However, algorithms accelerated by previous recognition processors [1, 2] are limited to specific applications, therefore, each IoE device may require an application-specific accelerator. On the other hand, deep convolutional neural networks (CNNs) [3] are a promising machine-learning approach, showing state-of-the-art recognition accuracy in a wide variety of applications, including both image and audio recognition. This makes CNNs a suitable candidate for a universal recognition platform for IoE devices, as described in Fig. 14.6.1. Due to the computational complexity and significant memory requirements of CNNs, a microcontroller unit (MCU) typically used for IoE devices is incapable of producing a meaningful recognition result in an energy-efficient way. Hence, the implementation of an energy-efficient CNN processor is desired to realize intelligent IoE systems.

AB - Transmitting massive amounts of image and audio data acquired by Internet-of-Everything (IoE) devices to data center servers for intelligent recognition processes is impractical for energy reasons, requiring in-situ processing of such data. However, algorithms accelerated by previous recognition processors [1, 2] are limited to specific applications, therefore, each IoE device may require an application-specific accelerator. On the other hand, deep convolutional neural networks (CNNs) [3] are a promising machine-learning approach, showing state-of-the-art recognition accuracy in a wide variety of applications, including both image and audio recognition. This makes CNNs a suitable candidate for a universal recognition platform for IoE devices, as described in Fig. 14.6.1. Due to the computational complexity and significant memory requirements of CNNs, a microcontroller unit (MCU) typically used for IoE devices is incapable of producing a meaningful recognition result in an energy-efficient way. Hence, the implementation of an energy-efficient CNN processor is desired to realize intelligent IoE systems.

UR - http://www.scopus.com/inward/record.url?scp=84962847015&partnerID=8YFLogxK

U2 - 10.1109/ISSCC.2016.7418008

DO - 10.1109/ISSCC.2016.7418008

M3 - Conference contribution

AN - SCOPUS:84962847015

T3 - Digest of Technical Papers - IEEE International Solid-State Circuits Conference

SP - 264

EP - 265

BT - 2016 IEEE International Solid-State Circuits Conference, ISSCC 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 63rd IEEE International Solid-State Circuits Conference, ISSCC 2016

Y2 - 31 January 2016 through 4 February 2016

ER -

Sim J, Park JS, Kim M, Bae D, Choi Y, Kim LS. A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems. In 2016 IEEE International Solid-State Circuits Conference, ISSCC 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 264-265. 7418008. (Digest of Technical Papers - IEEE International Solid-State Circuits Conference). doi: 10.1109/ISSCC.2016.7418008

A 1.42TOPS/W deep convolutional neural network recognition processor for intelligent IoE systems

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this