ABSTRACT
Learning in artificial neural networks is a process by which experience arising from exposure to measurements of empirical phenomena is converted to knowledge, embodied in network weights. This process can be viewed formally as statistical estimation of the parameters of a parametrized probability model. In this chapter we exploit this formal viewpoint to obtain a unified theory of learning in artificial neural networks. The theory is sufficiently general to encompass both supervised and unsupervised learning in either feedforward or recurrent networks.
