Reinforcement learning algorithm with network extension for pulse neural network

Koichiro Takita, Yuko Osana, Masafumi Hagiwara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a new hierarchical pulse neural network and its reinforcement learning algorithm with network extension. The proposed pulse neural network has three layers, and all of the neurons are pulse neurons. This network learns relations between input pulse sequences and the desired outputs by updating connection weights and by adding neurons dynamically. We carried out the computer simulation to confirm the performance of the proposed algorithm.

Original languageEnglish
Title of host publicationProceedings of the IEEE International Conference on Systems, Man and Cybernetics
PublisherIEEE
Pages2586-2591
Number of pages6
Volume4
Publication statusPublished - 2000
Event2000 IEEE International Conference on Systems, Man and Cybernetics - Nashville, TN, USA
Duration: 2000 Oct 82000 Oct 11

Other

Other2000 IEEE International Conference on Systems, Man and Cybernetics
CityNashville, TN, USA
Period00/10/800/10/11

Fingerprint

Reinforcement learning
Learning algorithms
Neurons
Neural networks
Computer simulation

ASJC Scopus subject areas

  • Hardware and Architecture
  • Control and Systems Engineering

Cite this

Takita, K., Osana, Y., & Hagiwara, M. (2000). Reinforcement learning algorithm with network extension for pulse neural network. In Proceedings of the IEEE International Conference on Systems, Man and Cybernetics (Vol. 4, pp. 2586-2591). IEEE.

Reinforcement learning algorithm with network extension for pulse neural network. / Takita, Koichiro; Osana, Yuko; Hagiwara, Masafumi.

Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. Vol. 4 IEEE, 2000. p. 2586-2591.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Takita, K, Osana, Y & Hagiwara, M 2000, Reinforcement learning algorithm with network extension for pulse neural network. in Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. vol. 4, IEEE, pp. 2586-2591, 2000 IEEE International Conference on Systems, Man and Cybernetics, Nashville, TN, USA, 00/10/8.
Takita K, Osana Y, Hagiwara M. Reinforcement learning algorithm with network extension for pulse neural network. In Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. Vol. 4. IEEE. 2000. p. 2586-2591
Takita, Koichiro ; Osana, Yuko ; Hagiwara, Masafumi. / Reinforcement learning algorithm with network extension for pulse neural network. Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. Vol. 4 IEEE, 2000. pp. 2586-2591
@inproceedings{fe06f36e017844f7a07035698c5b9ffb,
title = "Reinforcement learning algorithm with network extension for pulse neural network",
abstract = "In this paper, we propose a new hierarchical pulse neural network and its reinforcement learning algorithm with network extension. The proposed pulse neural network has three layers, and all of the neurons are pulse neurons. This network learns relations between input pulse sequences and the desired outputs by updating connection weights and by adding neurons dynamically. We carried out the computer simulation to confirm the performance of the proposed algorithm.",
author = "Koichiro Takita and Yuko Osana and Masafumi Hagiwara",
year = "2000",
language = "English",
volume = "4",
pages = "2586--2591",
booktitle = "Proceedings of the IEEE International Conference on Systems, Man and Cybernetics",
publisher = "IEEE",

}

TY - GEN

T1 - Reinforcement learning algorithm with network extension for pulse neural network

AU - Takita, Koichiro

AU - Osana, Yuko

AU - Hagiwara, Masafumi

PY - 2000

Y1 - 2000

N2 - In this paper, we propose a new hierarchical pulse neural network and its reinforcement learning algorithm with network extension. The proposed pulse neural network has three layers, and all of the neurons are pulse neurons. This network learns relations between input pulse sequences and the desired outputs by updating connection weights and by adding neurons dynamically. We carried out the computer simulation to confirm the performance of the proposed algorithm.

AB - In this paper, we propose a new hierarchical pulse neural network and its reinforcement learning algorithm with network extension. The proposed pulse neural network has three layers, and all of the neurons are pulse neurons. This network learns relations between input pulse sequences and the desired outputs by updating connection weights and by adding neurons dynamically. We carried out the computer simulation to confirm the performance of the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=0034511471&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034511471&partnerID=8YFLogxK

M3 - Conference contribution

VL - 4

SP - 2586

EP - 2591

BT - Proceedings of the IEEE International Conference on Systems, Man and Cybernetics

PB - IEEE

ER -