Characteristics of small scale non-monotonic neuron networks having large potentiality for learning

Mitsunaga Kinjo; Shigeo Sato; Koji Nakajima

Characteristics of small scale non-monotonic neuron networks having large potentiality for learning

Mitsunaga Kinjo, Shigeo Sato, Koji Nakajima

Information Devices Division

Research output: Contribution to conference › Paper › peer-review

1 Citation (Scopus)

Abstract

In this paper, we report a study on learning ability of a Deterministic Boltzmann Machine (DBM) with neurons which have a non-monotonic activation function. We use an end-cut-off-type function with a threshold parameter `θ' as the non-monotonic function. Numerical simulations of nonlinear problems, such as the 2-Parity problem and the 4-Parity problem, show that the DBM network with non-monotonic neurons has higher learning ability compared to the network with monotonic neurons.

Original language	English
Pages	171-174
Number of pages	4
Publication status	Published - 2000 Jan 1
Event	International Joint Conference on Neural Networks (IJCNN'2000) - Como, Italy Duration: 2000 Jul 24 → 2000 Jul 27

Other

Other	International Joint Conference on Neural Networks (IJCNN'2000)
City	Como, Italy
Period	00/7/24 → 00/7/27

ASJC Scopus subject areas

Software

Access to Document

Fingerprint Dive into the research topics of 'Characteristics of small scale non-monotonic neuron networks having large potentiality for learning'. Together they form a unique fingerprint.

Cite this

@conference{b49503747c4243cf969062693e9a5b7e,

title = "Characteristics of small scale non-monotonic neuron networks having large potentiality for learning",

abstract = "In this paper, we report a study on learning ability of a Deterministic Boltzmann Machine (DBM) with neurons which have a non-monotonic activation function. We use an end-cut-off-type function with a threshold parameter `θ' as the non-monotonic function. Numerical simulations of nonlinear problems, such as the 2-Parity problem and the 4-Parity problem, show that the DBM network with non-monotonic neurons has higher learning ability compared to the network with monotonic neurons.",

author = "Mitsunaga Kinjo and Shigeo Sato and Koji Nakajima",

year = "2000",

month = jan,

day = "1",

language = "English",

pages = "171--174",

note = "International Joint Conference on Neural Networks (IJCNN'2000) ; Conference date: 24-07-2000 Through 27-07-2000",

}

TY - CONF

T1 - Characteristics of small scale non-monotonic neuron networks having large potentiality for learning

AU - Kinjo, Mitsunaga

AU - Sato, Shigeo

AU - Nakajima, Koji

PY - 2000/1/1

Y1 - 2000/1/1

N2 - In this paper, we report a study on learning ability of a Deterministic Boltzmann Machine (DBM) with neurons which have a non-monotonic activation function. We use an end-cut-off-type function with a threshold parameter `θ' as the non-monotonic function. Numerical simulations of nonlinear problems, such as the 2-Parity problem and the 4-Parity problem, show that the DBM network with non-monotonic neurons has higher learning ability compared to the network with monotonic neurons.

AB - In this paper, we report a study on learning ability of a Deterministic Boltzmann Machine (DBM) with neurons which have a non-monotonic activation function. We use an end-cut-off-type function with a threshold parameter `θ' as the non-monotonic function. Numerical simulations of nonlinear problems, such as the 2-Parity problem and the 4-Parity problem, show that the DBM network with non-monotonic neurons has higher learning ability compared to the network with monotonic neurons.

UR - http://www.scopus.com/inward/record.url?scp=0033683810&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033683810&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0033683810

SP - 171

EP - 174

T2 - International Joint Conference on Neural Networks (IJCNN'2000)

Y2 - 24 July 2000 through 27 July 2000

ER -