Simultaneous design of feature extractor and pattern classifier using the minimum classification error training algorithm

Author

Paliwal, K.K. ; Bacchiani, M. ; Sagisaka, Y.

Author_Institution

ATR Interpreting Telephony Res. Labs., Kyoto, Japan

fYear

1995

fDate

31 Aug-2 Sep 1995

Firstpage

67

Lastpage

76

Abstract

Recently, a minimum classification error training algorithm has been proposed for minimizing the misclassification probability based on a given set of training samples using a generalized probabilistic descent method. This algorithm is a type of discriminative learning algorithm, but it approaches the objective of minimum classification error in a more direct manner than the conventional discriminative training algorithms. We apply this algorithm for simultaneous design of feature extractor and pattern classifier, and demonstrate some of its properties and advantages

Keywords

error statistics; feature extraction; learning (artificial intelligence); optimisation; pattern classification; probability; speech recognition; convergence; discriminative learning algorithm; feature extraction; minimum classification error training; multiple speaker vowel recognition; pattern classifier; probabilistic descent method; speech recognition; Algorithm design and analysis; Cepstral analysis; Classification algorithms; Feature extraction; Filter bank; Hidden Markov models; Pattern classification; Pattern recognition; Speech recognition; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing [1995] V. Proceedings of the 1995 IEEE Workshop

Conference_Location

Cambridge, MA

Print_ISBN

0-7803-2739-X

Type

conf

DOI

10.1109/NNSP.1995.514880

Filename

514880