Self-Organized Neural Networks

Abstract

As introduced in previous chapters, adaptation or learning is the main feature of artificial neural networks. In Chapter 6 on feedforward network, the focus was on various algorithms for supervised learning in which learning process is carried out with training data which is consisting of specific inputs and the desired outputs or intermediate states of the network. The training data is given by the supervisor of the network or an external signal source. For feedback networks (non-adaptive nets) discussed in Chapter 7, the weights are predetermined by the network designer for solving a specific problem. In this chapter we briefly introduce the concept of self-organized learning or unsupervised learning.