Japanese / English

Detail of Publication

Text Language English
Authors Masakazu Iwamura, Shinichiro Omachi, and Hirotomo Aso
Title On the Bias of Predictive Distribution in Pattern Recognition
Journal Systems and Computers in Japan
Vol. 36
No. 5
Pages pp.45-54
Publisher John Wiley & Sons, Inc.
Reviewed or not Not reviewed
Month & Year May 2005
Abstract To estimate distribution from training samples, the maximum likelihood estimation treats the unknown parameter of the distribution as a constant. On the other hand, Bayesian estimation treats the parameter as a random variable. In pattern recognition, Bayesian estimation has been known to improve recognition accuracy. However, it was pointed out that the Bayesian estimation is not effective due to the bias of the likelihood when sample sizes of classes are not the same. In this paper, we show recognition accuracy is improved by modifying the bias of the likelihood when sample sizes are not the same. This indicates that the cause of the ineffectiveness is the bias. We derive the formula of the bias of Geisser's predictive distribution without any approximation, and show a way of modification of the bias of the likelihood. We confirm effectiveness of the proposed method in experiments. In addition, the derived formula gives the theoretical background of a known empirical knowledge.
Back to list