An ANN Learning Algorithm Based on Hierarchical Clustering of Training Data

An ANN Learning Algorithm Based on Hierarchical Clustering of Training Data Tatsuya Uno, Student-Member, Seiichi Koakutsu, Member, Hironori Hirata, Member (Chiba University) We propose a new ANN learning algorithm based on hierarchical clustering of training data. The proposed algorithm first constructs a tree of sub-learning problems by hiearchically clustering given learning patterns in a bottom-up manner and decides a corresponding network structure. The proposed algorithm trains the whole network giving teacher signals of the original learning problem to the output units, and trains subnetworks giving teacher signals of the divided sub-learning problems to the hidden units simultaneously. The hidden units which learn sub-learning problems become feature detectors, which promote the learning of the original learning problem. We demonstrate the advantages of our learning algorithm by solving N-bits parity problems, a non-liner function approximation, iris classification problem, and two-spirals problem. Experimental results show that our learning algorithm obtains better solutions than the standard back-propagation algorithms and one of constructive algorithms in terms of the learning speed and the convergence rate.