Abstract
The corner classification approach to neural network training has the excellent capability ofprescriptive learning, where the network weights areprescribed merely by inspection of the training samples. This technique is extremely fast compared to other conventional training techniques such as backpropagation. However, the versions described hitherto have been sensitive to the choice of the radius of generalization. We present here a new and improved corner classification technique that retains the prescriptive learning capability and gives excellent generalization performance. This algorithm could be the basis of the recently introduced neuroscientific notion of “working memory.”
Similar content being viewed by others
References
Baddelely, A., and Della Sala, S., Working memory and executive control,Philosophical Trans. of the Royal Society of London, 351, 1996, 1397–1406.
Gazzaniga, M. S. (editor),The Cognitive Neuroscience, M.I.T. Press, Cambridge, MA, 1995.
Kak, S. C., New training algorithm in feedforward neural networks,First International Conference on Fuzzy Theory and Technology, Durham, NC, October 1992. Also in Wang, P. P. (editor),Advances in Fuzzy Theory and Technologies, Bookwright Press, Durham, NC, 1993.
Kak, S. C., On training feedforward neural networks,Pramana J. Physics, 40, 1993, 35–42.
Kak, S. C., New algorithms for training feedforward neural networks.Pattern Recognition Letters, 15, 1994, 295–298.
Kak, S. C., On generalization by neural networks,First International Conference on Computational Intelligence and Neuroscience, NC, 1995.
Kak, S. C., and Pastor, J., Neural networks and methods for training neural networks, U.S. Patent No. 5,426,721, June 20, 1995.
Looney, C. G.,Pattern Recognition Using Neural Networks, Oxford University Press, Oxford, 1997.
Porter, W. A., and Abouali, A. H., On neural network design, Part I: Using the MVQ algorithm,Circuits, Systems, and Signal Processing, 1997.
Porter, W. A., and Abouali, A. H., On neural network design, Part II: Inhibition and the output map,Circuits, Systems, and Signal Processing, 1997.
Raina, P., Comparison of learning and generalization capabilities of the Kak and the backpropagation algorithms,Information Sciences, 81, 1994, 261–274.
Sternberg, M. J. E. (editor),Protein Structure Prediction, Oxford University Press, Oxford, 1996.
Tang, K. W., and Kak, S. C., Corner classification that allows inhibitory output weights,Second International Conference on Computational Intelligence and Neuroscience, NC, March 1997.
Wickelgren, J., Getting a grasp on working memory,Science, 275, 1997, 1580–1582.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Tang, KW., Kak, S.C. A new corner classification approach to neural network training. Circuits Systems and Signal Process 17, 459–469 (1998). https://doi.org/10.1007/BF01201502
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF01201502