Modeling anticipatory behavior with self-organizing neural networks
Matti Pöllä and Timo Honkela
In: Computing Anticipatory Systems (CASYS'05), 8-13 Aug 2005, Liege, Belgium.

Abstract

A vital mechanism of high-level natural cognitive systems is the anticipatory capability of making decisions based on predicted events in the future. While in some cases the performance of computational cognitive systems can be improved with anticipatory behavior, it has been shown that for many cognitive tasks anticipation is mandatory. In this paper, we review the use of self-organizing artificial neural networks in constructing the state-space model of a state anticipatory system. The biologically inspired Self-Organizing Map (SOM) and its topologically dynamic variants such as the Growing Neural Gas are discussed together with illustrative examples of their performance.

[Edit]