PRUNING WITH REPLACEMENT ON LIMITED RESOURCE ALLOCATING NETWORKS BY F-PROJECTIONS

The principle of F-projection, in sequential function estimation, prov ides a theoretical foundation for a class of gaussian radial basis fun ction networks known as the resource allocating networks (RAN). The ad hoc rules for adaptively changing the size of RAN architectures can b e justified from a geometric growth criterion defined in the function space. In this paper, we show that the same arguments can be used to a rrive at a pruning with replacement rule for RAN architectures with a limited number of units. We illustrate the algorithm on the laser time series prediction problem of the Santa Fe competition and show that r esults similar to those of the winners of the competition can be obtai ned with pruning and replacement.