Hierarchical ensemble of neural networks
Procheta Poddar, P. V. Rao · 2002
The estimation of the a posteriori probability p(c/sub k/ mod x) given the state conditional probability distribution p(x mod c/sub k/) and a priori probability p(c/sub k/) is the central theme in the Bayesian approach to the pattern classification problem. The a posteriori probability can be expressed in a product form p(g/sub m/ mod x)p (c/sub k/ mod xg/sub m/). A classification scheme using a hierarchical ensemble of multilayer perceptrons (MLPs) is proposed based on this idea. This architecture is shown to be equivalent, in principle, to a single-stage MLP classifier. The advantages of the hierarchical ensemble of classifiers become apparent in practice where the probability estimates are computed from a finite set of samples in a finite time with a particular algorithm. With respect to given performance criteria, such as classification accuracy over a disjoint test set, a hierarchical ensemble performs better than an equivalent single-stage classifier, given a limited amount of resources in terms of input data and learning time. Experiments on vowel classification using a hierarchical scheme show these advantages over a single-stage classifier.>