Implementation of multi-layer neural network system for neuromorphic hardware architecture

Wookyung Sun; Junhee Park; Sumin Jo; Jungwon Lee; Hyungsoon Shin

doi:10.23919/ELINFOCOM.2019.8706456

Implementation of multi-layer neural network system for neuromorphic hardware architecture

Wookyung Sun, Junhee Park, Sumin Jo, Jungwon Lee, Hyungsoon Shin

Electronic and Electrical Engineering

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

Abstract

We propose a new neuromorphic hardware system that is optimized to implement a multi-layer guide training algorithm, which is a kind of reinforcement training algorithm. To consider the hardware implementation, we apply the guide training algorithm that is simple and very suitable for memristor synapse. The system is modeled using Simulink and the accuracy of the system is verified by classifying 'T', 'X', and 'V' in 3x3 letter image. The target image of hidden layer is set to the inverted image of the input image. Using this proposed system architecture, the reinforcement learning in multi-layer can be easily implemented in hardware.

Original language	English
Title of host publication	ICEIC 2019 - International Conference on Electronics, Information, and Communication
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9788995004449
DOIs	https://doi.org/10.23919/ELINFOCOM.2019.8706456
State	Published - 3 May 2019
Event	18th International Conference on Electronics, Information, and Communication, ICEIC 2019 - Auckland, New Zealand Duration: 22 Jan 2019 → 25 Jan 2019

Publication series

Name	ICEIC 2019 - International Conference on Electronics, Information, and Communication

Conference

Conference	18th International Conference on Electronics, Information, and Communication, ICEIC 2019
Country/Territory	New Zealand
City	Auckland
Period	22/01/19 → 25/01/19

Bibliographical note

Publisher Copyright:
© 2019 Institute of Electronics and Information Engineers (IEIE).

Keywords

Guide training algorithm
Hardware architecture
Multi-layer
Neral network
Reinforcement learning

Access to Document

10.23919/ELINFOCOM.2019.8706456

Cite this

Sun, W., Park, J., Jo, S., Lee, J., & Shin, H. (2019). Implementation of multi-layer neural network system for neuromorphic hardware architecture. In ICEIC 2019 - International Conference on Electronics, Information, and Communication Article 8706456 (ICEIC 2019 - International Conference on Electronics, Information, and Communication). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/ELINFOCOM.2019.8706456

Sun, Wookyung ; Park, Junhee ; Jo, Sumin et al. / Implementation of multi-layer neural network system for neuromorphic hardware architecture. ICEIC 2019 - International Conference on Electronics, Information, and Communication. Institute of Electrical and Electronics Engineers Inc., 2019. (ICEIC 2019 - International Conference on Electronics, Information, and Communication).

@inproceedings{a8a636193c7a4f278e9b0f49e669bb82,

title = "Implementation of multi-layer neural network system for neuromorphic hardware architecture",

abstract = "We propose a new neuromorphic hardware system that is optimized to implement a multi-layer guide training algorithm, which is a kind of reinforcement training algorithm. To consider the hardware implementation, we apply the guide training algorithm that is simple and very suitable for memristor synapse. The system is modeled using Simulink and the accuracy of the system is verified by classifying 'T', 'X', and 'V' in 3x3 letter image. The target image of hidden layer is set to the inverted image of the input image. Using this proposed system architecture, the reinforcement learning in multi-layer can be easily implemented in hardware.",

keywords = "Guide training algorithm, Hardware architecture, Multi-layer, Neral network, Reinforcement learning",

author = "Wookyung Sun and Junhee Park and Sumin Jo and Jungwon Lee and Hyungsoon Shin",

note = "Publisher Copyright: {\textcopyright} 2019 Institute of Electronics and Information Engineers (IEIE).; 18th International Conference on Electronics, Information, and Communication, ICEIC 2019 ; Conference date: 22-01-2019 Through 25-01-2019",

year = "2019",

month = may,

day = "3",

doi = "10.23919/ELINFOCOM.2019.8706456",

language = "English",

series = "ICEIC 2019 - International Conference on Electronics, Information, and Communication",

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

booktitle = "ICEIC 2019 - International Conference on Electronics, Information, and Communication",

}

Sun, W, Park, J, Jo, S, Lee, J & Shin, H 2019, Implementation of multi-layer neural network system for neuromorphic hardware architecture. in ICEIC 2019 - International Conference on Electronics, Information, and Communication., 8706456, ICEIC 2019 - International Conference on Electronics, Information, and Communication, Institute of Electrical and Electronics Engineers Inc., 18th International Conference on Electronics, Information, and Communication, ICEIC 2019, Auckland, New Zealand, 22/01/19. https://doi.org/10.23919/ELINFOCOM.2019.8706456

Implementation of multi-layer neural network system for neuromorphic hardware architecture. / Sun, Wookyung; Park, Junhee; Jo, Sumin et al.
ICEIC 2019 - International Conference on Electronics, Information, and Communication. Institute of Electrical and Electronics Engineers Inc., 2019. 8706456 (ICEIC 2019 - International Conference on Electronics, Information, and Communication).

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

TY - GEN

T1 - Implementation of multi-layer neural network system for neuromorphic hardware architecture

AU - Sun, Wookyung

AU - Park, Junhee

AU - Jo, Sumin

AU - Lee, Jungwon

AU - Shin, Hyungsoon

PY - 2019/5/3

Y1 - 2019/5/3

N2 - We propose a new neuromorphic hardware system that is optimized to implement a multi-layer guide training algorithm, which is a kind of reinforcement training algorithm. To consider the hardware implementation, we apply the guide training algorithm that is simple and very suitable for memristor synapse. The system is modeled using Simulink and the accuracy of the system is verified by classifying 'T', 'X', and 'V' in 3x3 letter image. The target image of hidden layer is set to the inverted image of the input image. Using this proposed system architecture, the reinforcement learning in multi-layer can be easily implemented in hardware.

AB - We propose a new neuromorphic hardware system that is optimized to implement a multi-layer guide training algorithm, which is a kind of reinforcement training algorithm. To consider the hardware implementation, we apply the guide training algorithm that is simple and very suitable for memristor synapse. The system is modeled using Simulink and the accuracy of the system is verified by classifying 'T', 'X', and 'V' in 3x3 letter image. The target image of hidden layer is set to the inverted image of the input image. Using this proposed system architecture, the reinforcement learning in multi-layer can be easily implemented in hardware.

KW - Guide training algorithm

KW - Hardware architecture

KW - Multi-layer

KW - Neral network

KW - Reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=85065878400&partnerID=8YFLogxK

U2 - 10.23919/ELINFOCOM.2019.8706456

DO - 10.23919/ELINFOCOM.2019.8706456

M3 - Conference contribution

AN - SCOPUS:85065878400

T3 - ICEIC 2019 - International Conference on Electronics, Information, and Communication

BT - ICEIC 2019 - International Conference on Electronics, Information, and Communication

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 18th International Conference on Electronics, Information, and Communication, ICEIC 2019

Y2 - 22 January 2019 through 25 January 2019

ER -

Sun W, Park J, Jo S, Lee J, Shin H. Implementation of multi-layer neural network system for neuromorphic hardware architecture. In ICEIC 2019 - International Conference on Electronics, Information, and Communication. Institute of Electrical and Electronics Engineers Inc. 2019. 8706456. (ICEIC 2019 - International Conference on Electronics, Information, and Communication). doi: 10.23919/ELINFOCOM.2019.8706456

Implementation of multi-layer neural network system for neuromorphic hardware architecture

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this