Continuous Optimization based-on Boosting Gaussian Mixture Mod

Authors:
Bin Lin;Xian-ji Wang;Run-tian Zhong;Zhen-quan Zhuang
Affiliations:
University of Science and Technology of China, Hefei, China;University of Science and Technology of China, Hefei, China;University of Science and Technology of China, Hefei, China;University of Science and Technology of China, Hefei, China
Venue:
ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 01
Year:
2006

Citing 0
Cited 3

Object classification by fusing SVMs and Gaussian mixtures

Pattern Recognition
Bayesian hybrid generative discriminative learning based on finite Liouville mixture models

Pattern Recognition
A review on probabilistic graphical models in evolutionary computation

Journal of Heuristics

Quantified Score

Hi-index	0.00

Visualization

Abstract

A new Estimation of Distribution Algorithm(EDA) based-on Gaussian Mixture Model (GMM) is proposed, in which boosting, an efficient ensemble learning method, is adopted to estimate GMM. By boosting simple GMM with two components, it has the ability of learning the model structure and parameters automatically without any requirement for prior knowledge. Moreover, since boosting can be viewed as a gradient search for a good fit of some objective in function space, the new EDA is time efficient. A set of experiments is implemented to evaluate the efficiency and performance of the new algorithm. The results show that, with a relatively smaller population and less number of generations, the new algorithm can perform as well as compared EDAs in optimizing multimodal functions.