Greedy Edge-Wise Training of Resistive Switch Arrays

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

Abstract

A technical challenge of machine learning based on artificial neural network is large-scale multiply-accumulate (MAC) operation that is costly. The larger the network, the more MAC operations are required for inference as well as training. As an alternative to this conventional digital MAC operation, a resistive switching memory array realizes analog MAC operations in a fully parallel manner. Algorithms for training such an array are mainly taken from conventional machine learning algorithms such as backpropagation. They are also customized such that they are suitably implemented on the array. In this chapter, we address such customized machine learning algorithm as well as new algorithms that are barely based on the conventional machine learning. A particular focus will be placed on a recently proposed greedy edge-wise training algorithm that is suitable for resistive switching memory arrays.

Original languageEnglish
Title of host publicationSpringer Series in Advanced Microelectronics
PublisherSpringer Verlag
Pages177-190
Number of pages14
DOIs
StatePublished - 2020 Jan 1

Publication series

NameSpringer Series in Advanced Microelectronics
Volume63
ISSN (Print)1437-0387

Fingerprint

Learning systems
Switches
Learning algorithms
Data storage equipment
Backpropagation
Neural networks

Cite this

Jeong, D. S. (2020). Greedy Edge-Wise Training of Resistive Switch Arrays. In Springer Series in Advanced Microelectronics (pp. 177-190). (Springer Series in Advanced Microelectronics; Vol. 63). Springer Verlag. https://doi.org/10.1007/978-981-13-8379-3_7
Jeong, Doo Seok. / Greedy Edge-Wise Training of Resistive Switch Arrays. Springer Series in Advanced Microelectronics. Springer Verlag, 2020. pp. 177-190 (Springer Series in Advanced Microelectronics).
@inbook{12cf058e22f949aeb12983124930c81b,
title = "Greedy Edge-Wise Training of Resistive Switch Arrays",
abstract = "A technical challenge of machine learning based on artificial neural network is large-scale multiply-accumulate (MAC) operation that is costly. The larger the network, the more MAC operations are required for inference as well as training. As an alternative to this conventional digital MAC operation, a resistive switching memory array realizes analog MAC operations in a fully parallel manner. Algorithms for training such an array are mainly taken from conventional machine learning algorithms such as backpropagation. They are also customized such that they are suitably implemented on the array. In this chapter, we address such customized machine learning algorithm as well as new algorithms that are barely based on the conventional machine learning. A particular focus will be placed on a recently proposed greedy edge-wise training algorithm that is suitable for resistive switching memory arrays.",
author = "Jeong, {Doo Seok}",
year = "2020",
month = "1",
day = "1",
doi = "10.1007/978-981-13-8379-3_7",
language = "English",
series = "Springer Series in Advanced Microelectronics",
publisher = "Springer Verlag",
pages = "177--190",
booktitle = "Springer Series in Advanced Microelectronics",

}

Jeong, DS 2020, Greedy Edge-Wise Training of Resistive Switch Arrays. in Springer Series in Advanced Microelectronics. Springer Series in Advanced Microelectronics, vol. 63, Springer Verlag, pp. 177-190. https://doi.org/10.1007/978-981-13-8379-3_7

Greedy Edge-Wise Training of Resistive Switch Arrays. / Jeong, Doo Seok.

Springer Series in Advanced Microelectronics. Springer Verlag, 2020. p. 177-190 (Springer Series in Advanced Microelectronics; Vol. 63).

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

TY - CHAP

T1 - Greedy Edge-Wise Training of Resistive Switch Arrays

AU - Jeong, Doo Seok

PY - 2020/1/1

Y1 - 2020/1/1

N2 - A technical challenge of machine learning based on artificial neural network is large-scale multiply-accumulate (MAC) operation that is costly. The larger the network, the more MAC operations are required for inference as well as training. As an alternative to this conventional digital MAC operation, a resistive switching memory array realizes analog MAC operations in a fully parallel manner. Algorithms for training such an array are mainly taken from conventional machine learning algorithms such as backpropagation. They are also customized such that they are suitably implemented on the array. In this chapter, we address such customized machine learning algorithm as well as new algorithms that are barely based on the conventional machine learning. A particular focus will be placed on a recently proposed greedy edge-wise training algorithm that is suitable for resistive switching memory arrays.

AB - A technical challenge of machine learning based on artificial neural network is large-scale multiply-accumulate (MAC) operation that is costly. The larger the network, the more MAC operations are required for inference as well as training. As an alternative to this conventional digital MAC operation, a resistive switching memory array realizes analog MAC operations in a fully parallel manner. Algorithms for training such an array are mainly taken from conventional machine learning algorithms such as backpropagation. They are also customized such that they are suitably implemented on the array. In this chapter, we address such customized machine learning algorithm as well as new algorithms that are barely based on the conventional machine learning. A particular focus will be placed on a recently proposed greedy edge-wise training algorithm that is suitable for resistive switching memory arrays.

UR - http://www.scopus.com/inward/record.url?scp=85069770621&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-8379-3_7

DO - 10.1007/978-981-13-8379-3_7

M3 - Chapter

T3 - Springer Series in Advanced Microelectronics

SP - 177

EP - 190

BT - Springer Series in Advanced Microelectronics

PB - Springer Verlag

ER -

Jeong DS. Greedy Edge-Wise Training of Resistive Switch Arrays. In Springer Series in Advanced Microelectronics. Springer Verlag. 2020. p. 177-190. (Springer Series in Advanced Microelectronics). https://doi.org/10.1007/978-981-13-8379-3_7