Fast Learning Algorithms for Neural Networks

  • N. B. Karayiannis
  • A. N. Venetsanopoulos
Part of the The Springer International Series in Engineering and Computer Science book series (SECS, volume 209)


When neural networks regained popularity in the early eighties, the dominant trend among researchers was an eagerness to develop impressive neural network applications and then quickly produce and market neural network systems. This trend was mainly motivated, and also justified, by a desire to establish neural networks as a viable and realistic alternative for developing speech and image recognition systems, and trainable control devices. During the early stages of this revitalized interest in neural network research, the existing learning algorithms were satisfactory. As research has moved from state-of-the-art paradigms to real-world applications, the associated training time and computing requirements have become an increasingly important consideration in the comparison of neural networks with alternative, competing techniques. The availability of fast and efficient learning algorithms is crucial for the future evolution of this research field.


Neural Network Artificial Neural Network Learning Rate Back Propagation Synaptic Weight 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media New York 1993

Authors and Affiliations

  • N. B. Karayiannis
    • 1
  • A. N. Venetsanopoulos
    • 2
  1. 1.University of HoustonUSA
  2. 2.University of TorontoCanada

Personalised recommendations