A multi-net local learning framework for pattern recognition

Author

Dong, Jian-xiong ; Krzyzak, Adam ; Suen, C.Y.

Author_Institution

Centre of Pattern Recognition & Machine Intelligence, Concordia Univ., Montreal, Que., Canada

fYear

2001

fDate

6/23/1905 12:00:00 AM

Firstpage

328

Lastpage

332

Abstract

This paper proposes a general local learning framework to effectively alleviate the complexities of classifier design by means of "divide and conquer" principle and ensemble method. The learning framework consists of quantization layer and ensemble layer. After GLVQ and MLP are applied to the framework, the proposed method is tested on MNIST handwritten digit database. The obtained performance is very promising, an error rate with 0.99%, which is comparable to that of LeNet5, one of the best classifiers on this database. Further, in contrast to LeNet5, our method is especially suitable for a large-scale real-world classification problem

Keywords

character recognition; learning (artificial intelligence); neural nets; pattern classification; vector quantisation; GLVQ; MLP; classification; divide and conquer; ensemble layer; handwritten digit database; learning framework; local learning framework; neural networks; pattern classification; pattern recognition; quantization layer; training data; Bagging; Databases; Machine intelligence; Machine learning; Neural networks; Pattern recognition; Piecewise linear approximation; Piecewise linear techniques; Switches; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis and Recognition, 2001. Proceedings. Sixth International Conference on

Conference_Location

Seattle, WA

Print_ISBN

0-7695-1263-1

Type

conf

DOI

10.1109/ICDAR.2001.953808

Filename

953808