Time-Varying Neurocomputing: An Iterative Learning Perspective
This paper proposes a unified architecture of time-varying neural networks for implementing unknown time-varying mappings. The methodology of iterative learning is applied for the network training, and a modified iterative learning least squares algorithm is presented. Under the assumption of bounded input signals, convergence results of the proposed learning algorithm are given. In order to realize periodic mappings, periodic neural networks are characterized and a periodic learning algorithm is presented for training such neural networks.
KeywordsNeural networks least squares learning algorithms time-varying system identification
Unable to display preview. Download preview PDF.
- 2.Ham, F.M., Kostanic, I.: Principles of Neurocomputing for Science & Engineering. McGraw-Hill, New York (2001)Google Scholar
- 7.Sun, M.: Iterative Learning Neurocomputing. In: Proceedings of 2009 International Conference on Wireless Networks and Information Systems, Shanghai, China, December 28-29, pp. 158–161 (2009)Google Scholar