Learning techniques for structured networks
Date
1991ISBN
0-87942-565-2Publisher
Publ by American Automatic Control CouncilSource
Proceedings of the American Control ConferenceProceedings of the 1991 American Control Conference
Volume
3Pages
2413-2418Google Scholar check
Keyword(s):
Metadata
Show full item recordAbstract
A convergence analysis is presented for the training of structured networks. Since the learning techniques that are used in structured networks are the same as the ones used in training of neural networks, the issue of convergence is discussed not only from a numerical perspective but also as a means of deriving insight into connectionist learning. In the analysis, bounds are developed on the learning rate, under which exponential convergence of the weights to their correct values is proved for a class of matrix algebra problems that includes linear equation solving, matrix inversion, and Lyapunov equation solving.