An Heuristic Pattern Correction Scheme for GRNNs and Its Application to Speech Recognition

In an on-line learning environment where optimal recognition performance over the newly encountered patterns is required , a robust incremental learning procedure is necessary to re-conngure the entire neural network without aaecting the stored information. In this paper, an heuristic pattern correction scheme based upon an hierarchical data partitioning principle is proposed for digit word recognition. This scheme is based upon General Regression Neural Networks (GRNNs) with initial cen-troid vectors obtained by graph theoretic data-pruning methods. Simulation results show that the proposed scheme can perfectly correct the mis-classiied patterns and hence improves the gen-eralisation performance without aaecting the old information. Moreover, it is also established that the initial setting of Radial Basis Functions (RBFs) based upon graph theoretic data-pruning methods yields better performance than those obtained by k-means and Learning Vector Quantisation (LVQ) methods.