A Cascade System of Dynamic Binary Neural Networks and Learning of Periodic Orbit

Jungo MORIYASU; Toshimichi SAITO

doi:10.1587/transinf.2014OPP0011

A Cascade System of Dynamic Binary Neural Networks and Learning of Periodic Orbit

Jungo MORIYASU, Toshimichi SAITO

Full Text Views

0

Cite this

Summary :

This paper studies a cascade system of dynamic binary neural networks. The system is characterized by signum activation function, ternary connection parameters, and integer threshold parameters. As a fundamental learning problem, we consider storage and stabilization of one desired binary periodic orbit that corresponds to control signals of switching circuits. For the storage, we present a simple method based on the correlation learning. For the stabilization, we present a sparsification method based on the mutation operation in the genetic algorithm. Using the Gray-code-based return map, the storage and stability can be investigated. Performing numerical experiments, effectiveness of the learning method is confirmed.

Publication: IEICE TRANSACTIONS on Information Vol.E98-D No.9 pp.1622-1629

Publication Date: 2015/09/01

Publicized: 2015/06/22

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2014OPP0011

Type of Manuscript: Special Section PAPER (Special Section on Optimization and Learning Algorithms of Small Embedded Devices and Related Software/Hardware Implementation)

Category

Authors

Jungo MORIYASU
Hosei University
Toshimichi SAITO
Hosei University

Keyword

binary neural networks, deep learning, switching circuits

Cite this

Copy

Jungo MORIYASU, Toshimichi SAITO, "A Cascade System of Dynamic Binary Neural Networks and Learning of Periodic Orbit" in IEICE TRANSACTIONS on Information, vol. E98-D, no. 9, pp. 1622-1629, September 2015, doi: 10.1587/transinf.2014OPP0011.
Abstract: This paper studies a cascade system of dynamic binary neural networks. The system is characterized by signum activation function, ternary connection parameters, and integer threshold parameters. As a fundamental learning problem, we consider storage and stabilization of one desired binary periodic orbit that corresponds to control signals of switching circuits. For the storage, we present a simple method based on the correlation learning. For the stabilization, we present a sparsification method based on the mutation operation in the genetic algorithm. Using the Gray-code-based return map, the storage and stability can be investigated. Performing numerical experiments, effectiveness of the learning method is confirmed.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2014OPP0011/_p

Copy

@ARTICLE{e98-d_9_1622,
author={Jungo MORIYASU, Toshimichi SAITO, },
journal={IEICE TRANSACTIONS on Information},
title={A Cascade System of Dynamic Binary Neural Networks and Learning of Periodic Orbit},
year={2015},
volume={E98-D},
number={9},
pages={1622-1629},
abstract={This paper studies a cascade system of dynamic binary neural networks. The system is characterized by signum activation function, ternary connection parameters, and integer threshold parameters. As a fundamental learning problem, we consider storage and stabilization of one desired binary periodic orbit that corresponds to control signals of switching circuits. For the storage, we present a simple method based on the correlation learning. For the stabilization, we present a sparsification method based on the mutation operation in the genetic algorithm. Using the Gray-code-based return map, the storage and stability can be investigated. Performing numerical experiments, effectiveness of the learning method is confirmed.},
keywords={},
doi={10.1587/transinf.2014OPP0011},
ISSN={1745-1361},
month={September},}

Copy

TY - JOUR
TI - A Cascade System of Dynamic Binary Neural Networks and Learning of Periodic Orbit
T2 - IEICE TRANSACTIONS on Information
SP - 1622
EP - 1629
AU - Jungo MORIYASU
AU - Toshimichi SAITO
PY - 2015
DO - 10.1587/transinf.2014OPP0011
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2015
AB - This paper studies a cascade system of dynamic binary neural networks. The system is characterized by signum activation function, ternary connection parameters, and integer threshold parameters. As a fundamental learning problem, we consider storage and stabilization of one desired binary periodic orbit that corresponds to control signals of switching circuits. For the storage, we present a simple method based on the correlation learning. For the stabilization, we present a sparsification method based on the mutation operation in the genetic algorithm. Using the Gray-code-based return map, the storage and stability can be investigated. Performing numerical experiments, effectiveness of the learning method is confirmed.
ER -