Randomness in generalization ability: a source to improve it?

Abstract

The question of generalization ability of artificial neural networks is of great interest for both theoretical understanding and practical use. This paper reports our observations about randomness in generalization ability of feedforward artificial neural networks (FFANNs). A novel method for measuring generalization ability is defined. This definition can be used to identify degree of randomness in generalization ability of learning systems. If an FFANN architecture shows randomness in generalization ability for a given problem then multiple networks can be used to improve it. We have developed a model, called voting model, for predicting generalization ability of multiple networks. It has been shown that if correct classification probability of a single network is greater than half, then as the number of networks is increased so does the generalization ability. >