Abstract
This paper points out some drawbacks and proposes some modifications to the conventional layer-by-layer BP algorithm. In particular, we present a new perspective to the learning rate, which is to use a heuristic rule to define the learning rate so as to update the weights. Meanwhile, to pull the algorithm out of saturation area and prevent it from converging to a local minimum, a momentum term is introduced to the former algorithm. And finally the effectiveness and efficiency of the proposed method are demonstrated by two benchmark examples.
This work was supported by the National Natural Science Foundation of China (Nos.60472111 and 60405002), and RGC Project No.CUHK 4170/04E, RGC Project No. CUHK4205/04E and UGC Project No.AoE/E-01/99.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Huang, D.S., Ip, H.H.S., Chi, Z.: A Neural Root Finder of Polynomials Based on Root Moments. Neural Computation 16(8), 1721–1762 (2004)
Huang, D.-S., Ip, H.H.S., Law, K.C.K., Chi, Z.: Zeroing Polynomials Using Modified Constrained Neural Network Approach. IEEE Trans. on Neural Networks 16(3), 721–732 (2005)
Huang, D.S.: A Constructive Approach for Finding Arbitrary Roots of Polynomials by Neural Networks. IEEE Transactions on Neural Networks 15(2), 477–491 (2004)
Huang, D.S.: Systematic Theory of Neural Networks for Pattern Recognition. Publishing House of Electronic Industry of China, Beijing (1996)
Rumelhart, D.E., McClelland, J.L.: Parallel Distributed Processing. MIT Press, Cambridge (1986)
Huang, D.S., Ip, H.H.S., Chi, Z.: A Neural Root Finder of Polynomials Based on Root Moments. Neural Computation 16(8), 1721–1762 (2004)
Huang, D.S.: The Local Minima Free Condition of Feedforward Neural Networks for Outer-supervised Learning. IEEE Trans. on Systems, Man and Cybernetics 28B(3), 477–480 (1998)
Huang, D.S.: Radial Basis Probabilistic Neural Networks: Model and Application. International Journal of Pattern Recognition and Artificial Intelligence 13(7), 1083–1101 (1999)
Oh, S.-H., Lee, S.-Y.: A New Error Function at Hidden Layers for Fast Training of Multilayer Perceptrons. IEEE Trans. Neural Networks 10, 960–964 (1999)
Ergezinger, S., Thomsen, E.: An Accelerated Learning Algorithm for Multilayer Pereceptrons: Optimization Layer by Layer. IEEE Trans. Neural Networks 6, 31–42 (1995)
Wang, G.-J., Chen, C.-C.: A fast Multilayer Neural Networks Training Algorithm Based on the Layer-by-layer Optimizing Procedures. IEEE Trans. Neural Networks 7, 768–775 (1996)
van Milligen, B.P., Tribaldos, V., Jimenez, J.A., Santa Cruz, C.: Comments on: An Accelerated Algorithm for Multilayer Perceptrons: Optimization Layer by Layer. IEEE Trans. Neural Networks 9, 339–341 (1998)
van Ooyen, Nienhuis, B.: Improving the Convergence of the Backpropagation Algorithm. Neural Networks 5, 465–471 (1992)
Huang, D.S.: The Bottleneck Behavior in Linear Feedforward Neural Network Classifiers and Their Breakthrough. Journal of Computer Science and Technology 14(1), 34–43 (1999)
Oh, S.-H.: Improving the Error Backpropagation Algorithm with a Modified Error Function. IEEE Trans. Neural Networks 8, 799–803 (1997)
Hagen, M.T., Demuth, H.B.: Neural Network Design. PWS publishing company, USA (1996)
LeCun, Y.A., Bottou, L., Orr, G.B., Müller, K.-R.: Efficient Backprop. In: Orr, G.B., Müller, K.-R. (eds.) NIPS-WS 1996. LNCS, vol. 1524, pp. 9–50. Springer, Heidelberg (1998)
Ham, F.M., Kostanic, I.: Principles of Neuaocomputing for Science and Engineering. McGraw-Hill Companies, Inc., USA (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Li, XQ., Han, F., Lok, TM., Lyu, M.R., Huang, GB. (2005). Improvements to the Conventional Layer-by-Layer BP Algorithm. In: Huang, DS., Zhang, XP., Huang, GB. (eds) Advances in Intelligent Computing. ICIC 2005. Lecture Notes in Computer Science, vol 3645. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11538356_20
Download citation
DOI: https://doi.org/10.1007/11538356_20
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-28227-3
Online ISBN: 978-3-540-31907-8
eBook Packages: Computer ScienceComputer Science (R0)