Design of Pattern Classifiers with the Updating Property Using Stochastic Approximation Techniques

Sik-Sang Yau, J. M. Schumpert

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

A nonparametric training procedure for finding the optimal weights of the discriminant functions of a pattern classifier in any optimization criterion, expressible as a convex function from an arbitrary sequence of sample patterns, is proposed. This design procedure is based on the stochastic approximation technique, and has the updating property because it processes the sample patterns whenever they become available. This procedure is used to find the optimal weights for the least-mean-square error criterion, and is shown to require very simple computation which leads to simple im-plementation. Both two-category and multi-category cases are considered, and an acceleration scheme to increase the rate of convergence for the training procedure is also presented. These results are demonstrated by examples.

Original languageEnglish (US)
Pages (from-to)861-872
Number of pages12
JournalIEEE Transactions on Computers
VolumeC-17
Issue number9
DOIs
StatePublished - 1968
Externally publishedYes

Fingerprint

Stochastic Approximation
Updating
Classifiers
Classifier
Mean square error
Discriminant Function
Least Mean Square
Convex function
Rate of Convergence
Design
Optimization
Arbitrary
Training

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Hardware and Architecture
  • Software
  • Theoretical Computer Science

Cite this

Design of Pattern Classifiers with the Updating Property Using Stochastic Approximation Techniques. / Yau, Sik-Sang; Schumpert, J. M.

In: IEEE Transactions on Computers, Vol. C-17, No. 9, 1968, p. 861-872.

Research output: Contribution to journalArticle

@article{01fbcd62cef54d4ab198f8fcca094f1e,
title = "Design of Pattern Classifiers with the Updating Property Using Stochastic Approximation Techniques",
abstract = "A nonparametric training procedure for finding the optimal weights of the discriminant functions of a pattern classifier in any optimization criterion, expressible as a convex function from an arbitrary sequence of sample patterns, is proposed. This design procedure is based on the stochastic approximation technique, and has the updating property because it processes the sample patterns whenever they become available. This procedure is used to find the optimal weights for the least-mean-square error criterion, and is shown to require very simple computation which leads to simple im-plementation. Both two-category and multi-category cases are considered, and an acceleration scheme to increase the rate of convergence for the training procedure is also presented. These results are demonstrated by examples.",
author = "Sik-Sang Yau and Schumpert, {J. M.}",
year = "1968",
doi = "10.1109/TC.1968.229146",
language = "English (US)",
volume = "C-17",
pages = "861--872",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "9",

}

TY - JOUR

T1 - Design of Pattern Classifiers with the Updating Property Using Stochastic Approximation Techniques

AU - Yau, Sik-Sang

AU - Schumpert, J. M.

PY - 1968

Y1 - 1968

N2 - A nonparametric training procedure for finding the optimal weights of the discriminant functions of a pattern classifier in any optimization criterion, expressible as a convex function from an arbitrary sequence of sample patterns, is proposed. This design procedure is based on the stochastic approximation technique, and has the updating property because it processes the sample patterns whenever they become available. This procedure is used to find the optimal weights for the least-mean-square error criterion, and is shown to require very simple computation which leads to simple im-plementation. Both two-category and multi-category cases are considered, and an acceleration scheme to increase the rate of convergence for the training procedure is also presented. These results are demonstrated by examples.

AB - A nonparametric training procedure for finding the optimal weights of the discriminant functions of a pattern classifier in any optimization criterion, expressible as a convex function from an arbitrary sequence of sample patterns, is proposed. This design procedure is based on the stochastic approximation technique, and has the updating property because it processes the sample patterns whenever they become available. This procedure is used to find the optimal weights for the least-mean-square error criterion, and is shown to require very simple computation which leads to simple im-plementation. Both two-category and multi-category cases are considered, and an acceleration scheme to increase the rate of convergence for the training procedure is also presented. These results are demonstrated by examples.

UR - http://www.scopus.com/inward/record.url?scp=84933800527&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84933800527&partnerID=8YFLogxK

U2 - 10.1109/TC.1968.229146

DO - 10.1109/TC.1968.229146

M3 - Article

AN - SCOPUS:84933800527

VL - C-17

SP - 861

EP - 872

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 9

ER -