New Learning Rules for Three-layered Feed-forward Neural Networks based on a General Learning Schema

Christina Klüver, Jürgen Klüver

2014

Abstract

We propose a general schema for learning rules in neural networks, the General Enforcing Rule Schema (GERS), from which we infer new simplified learning rules for in particular supervised learning of three-layered feed-forward networks, namely Enforcing Rule Supervised ERS and ERS2. These rules are comparatively simpler than the established rule of Backpropagation and their performance is at least equivalent. The new rules are compared with Backpropagation in different experiments; an additional comparison is performed with real data for the prediction of parcel delivering.

References

  1. Hebb, D.O.: The Organization of Behavior. Wiley, New York (1949)
  2. Klüver, C., Klüver, J.: Self-organized Learning by Self-Enforcing Networks. In: Rojas I, Joya G, Cabestany J (eds.): IWANN 2013, Part I, Lecture Notes in Computer Science Vol. 7902, Springer, Berlin Heidelberg (2013) 518-529
  3. Klüver, C., Klüver, J.: OSWI: A consulting system for pupils and prospective students on the basis of neural networks. In: AI & Society. Springer-Verlag, London (2014) DOI 10.1007/s00146-014-0542-y
  4. Orr, G., Müller, K.: Neural Networks: Tricks of the Trade, Springer, Berlin Heidelberg New York (1998)
  5. Patterson, D.: Artificial Neural Networks, Theory and Applications. Prentice Hall, New Jersey (1996)
  6. Seung, S.: Connectome. Houghton Mifflin Harcourt, Boston New York (2012)
  7. Thimm, G., Fiesler, E.: Optimal Setting of Weights, Learning Rate, and Gain. IDIAPResearch Report 04-1997 http://publications.idiap.ch/downloads/reports/1997/rr97-04.pdf (1997)
  8. Widrow, B., Greenblatt, A., Kim, Y., Park, D.: 2013. The No-Prop algorithm: A new learning algorithm for multilayer neural networks. Neural Networks 37. Elsevier Science Ltd, Oxford (2013) 182-188
Download


Paper Citation


in Harvard Style

Klüver C. and Klüver J. (2014). New Learning Rules for Three-layered Feed-forward Neural Networks based on a General Learning Schema . In Proceedings of the International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: ANNIIP, (ICINCO 2014) ISBN 978-989-758-041-3, pages 27-36. DOI: 10.5220/0005125600270036


in Bibtex Style

@conference{anniip14,
author={Christina Klüver and Jürgen Klüver},
title={New Learning Rules for Three-layered Feed-forward Neural Networks based on a General Learning Schema},
booktitle={Proceedings of the International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: ANNIIP, (ICINCO 2014)},
year={2014},
pages={27-36},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005125600270036},
isbn={978-989-758-041-3},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: ANNIIP, (ICINCO 2014)
TI - New Learning Rules for Three-layered Feed-forward Neural Networks based on a General Learning Schema
SN - 978-989-758-041-3
AU - Klüver C.
AU - Klüver J.
PY - 2014
SP - 27
EP - 36
DO - 10.5220/0005125600270036