Improved RAN sequential prediction using orthogonal techniques

A new learning strategy for time-series prediction using radial basis function (RBF) networks is introduced. Its potential is examined in the particular case of the resource allocating network model, although the same ideas could apply to extend any other procedure. In the early stages of learning, addition of successive new groups of RBFs provides an increased rate of convergence. At the same time, the optimum lag structure is determined using orthogonal techniques such as QR factorization and singular value decomposition (SVD). We claim that the same techniques can be applied to the pruning problem, and thus they are a useful tool for compaction of information. Our comparison with the original RAN algorithm shows a comparable error measure but much smaller-sized networks. The extra e!ort required by QR and SVD is balanced by the simplicity of only using least mean squares for the iterative parameter adaptation. 2001 Elsevier Science B.V. All rights reserved.