Pruning RBF networks with QLP decomposition
NN'07 Proceedings of the 8th Conference on 8th WSEAS International Conference on Neural Networks - Volume 8
NN'07 Proceedings of the 8th Conference on 8th WSEAS International Conference on Neural Networks - Volume 8
ACMOS'07 Proceedings of the 9th WSEAS international conference on Automatic control, modelling and simulation
Hi-index | 0.00 |
This paper presents a sequential decomposition of the data auto-correlation matrix using SVD, QR-method and QLP as pre-processors from the design of Wiener Filters. It is shown that this approach is effective for noise reduction by improving the SNR while reducing CPU time for a range of input signal lengths, filter length and noise variance. The theoretical and practical aspects of the proposed approach are introduced and compared to those obtained from simulation results.