Abstract
This article deals with back-propagation, a learning method for neural nets. It is shown, first, how the introduction of test cycles brings out a great improvement in the convergence rate and, second, that costly experiments used to adjust learning-relevant parameters could be dispensed with.
Preview
Unable to display preview. Download preview PDF.
References
Scott E. Fahlmann. Faster-learning variations on back-propagation: An empirical study. In David Touretzky, Geoffrey Hinton, and Terrence Sejnowski, editors, Proceedings of the 1988 Connectionist Models Summer School, pages 38–51, San Mateo, CA, 1989. Morgan Kaufmann Publishers.
Robert A. Jacobs. Increased rates of convergence through learning rate adaption. Neural Networks, I:295–307, 1988.
Alan H. Kramer and A. Sangiovanni-Vincentelli. Efficient parallel learning algorithms for neural networks. IEEE Computer, 1987.
Kevin J. Lang and Michael J. Witbrock. Learning to tell two spirals apart. In David Touretzky, Geoffrey Hinton, and Terrence Sejnowski, editors, Proceedings of the 1988 Connectionist Models Summer School, pages 56–59, San Mateo, CA, 1989. Morgan Kaufmann Publishers.
D. R. Rush and J. M. Salas. Improving the learning rate of backpropagation with the gradient reuse algorithm. In IEEE International Conference on Neural Networks, pages I-441, San Diego, CA, 1988. The Institute of Electrical and Electronic Engineers, Inc., IEEE San Diego Section and IEEE TAB Neural Network Committee.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Salomon, R. (1991). Improved convergence rate of back-propagation with dynamic adaption of the learning rate. In: Schwefel, HP., Männer, R. (eds) Parallel Problem Solving from Nature. PPSN 1990. Lecture Notes in Computer Science, vol 496. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0029763
Download citation
DOI: https://doi.org/10.1007/BFb0029763
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54148-6
Online ISBN: 978-3-540-70652-6
eBook Packages: Springer Book Archive