A neurocomputing framework

From methodologies to application

Research output: Contribution to journalArticle

Abstract

In this paper, we present a practical framework of methodologies for increasing the efficiency of the training process and improving the generalization capability of neural networks. The methodologies are devised for resolving problems of neural networks primarily in three aspects: learning, architecture, and data representation. For learning we present a rapid learning method based on Aitken's Δ2 process and a training schedule called selective reinforcement learning; for architecture, a two-stage classification scheme and a multiple network scheme; and for data representation, a data generation scheme with systematic noise and a preprocessing method by hidden Markov model. In order to investigate the behavior of neural network classifiers with the proposed methodologies, we designed and implemented neural networks for recognizing on-line handwriting characters obtained by an LCD tablet. Experimental results with a large set of on-line handwriting characters show the usefulness of the proposed methodologies.

Original languageEnglish
Pages (from-to)151-170
Number of pages20
JournalNeurocomputing
Volume13
Issue number2-4
DOIs
Publication statusPublished - 1996 Jan 1

Fingerprint

Learning
Handwriting
Neural networks
Reinforcement learning
Hidden Markov models
Liquid crystal displays
Tablets
Noise
Appointments and Schedules
Classifiers
Efficiency
Generalization (Psychology)
Reinforcement (Psychology)

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

@article{8f887f3d7c5648758aad7b5b5973c2a4,
title = "A neurocomputing framework: From methodologies to application",
abstract = "In this paper, we present a practical framework of methodologies for increasing the efficiency of the training process and improving the generalization capability of neural networks. The methodologies are devised for resolving problems of neural networks primarily in three aspects: learning, architecture, and data representation. For learning we present a rapid learning method based on Aitken's Δ2 process and a training schedule called selective reinforcement learning; for architecture, a two-stage classification scheme and a multiple network scheme; and for data representation, a data generation scheme with systematic noise and a preprocessing method by hidden Markov model. In order to investigate the behavior of neural network classifiers with the proposed methodologies, we designed and implemented neural networks for recognizing on-line handwriting characters obtained by an LCD tablet. Experimental results with a large set of on-line handwriting characters show the usefulness of the proposed methodologies.",
author = "Sung-Bae Cho",
year = "1996",
month = "1",
day = "1",
doi = "10.1016/0925-2312(95)00089-5",
language = "English",
volume = "13",
pages = "151--170",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",
number = "2-4",

}

A neurocomputing framework : From methodologies to application. / Cho, Sung-Bae.

In: Neurocomputing, Vol. 13, No. 2-4, 01.01.1996, p. 151-170.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A neurocomputing framework

T2 - From methodologies to application

AU - Cho, Sung-Bae

PY - 1996/1/1

Y1 - 1996/1/1

N2 - In this paper, we present a practical framework of methodologies for increasing the efficiency of the training process and improving the generalization capability of neural networks. The methodologies are devised for resolving problems of neural networks primarily in three aspects: learning, architecture, and data representation. For learning we present a rapid learning method based on Aitken's Δ2 process and a training schedule called selective reinforcement learning; for architecture, a two-stage classification scheme and a multiple network scheme; and for data representation, a data generation scheme with systematic noise and a preprocessing method by hidden Markov model. In order to investigate the behavior of neural network classifiers with the proposed methodologies, we designed and implemented neural networks for recognizing on-line handwriting characters obtained by an LCD tablet. Experimental results with a large set of on-line handwriting characters show the usefulness of the proposed methodologies.

AB - In this paper, we present a practical framework of methodologies for increasing the efficiency of the training process and improving the generalization capability of neural networks. The methodologies are devised for resolving problems of neural networks primarily in three aspects: learning, architecture, and data representation. For learning we present a rapid learning method based on Aitken's Δ2 process and a training schedule called selective reinforcement learning; for architecture, a two-stage classification scheme and a multiple network scheme; and for data representation, a data generation scheme with systematic noise and a preprocessing method by hidden Markov model. In order to investigate the behavior of neural network classifiers with the proposed methodologies, we designed and implemented neural networks for recognizing on-line handwriting characters obtained by an LCD tablet. Experimental results with a large set of on-line handwriting characters show the usefulness of the proposed methodologies.

UR - http://www.scopus.com/inward/record.url?scp=0030271847&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030271847&partnerID=8YFLogxK

U2 - 10.1016/0925-2312(95)00089-5

DO - 10.1016/0925-2312(95)00089-5

M3 - Article

VL - 13

SP - 151

EP - 170

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

IS - 2-4

ER -