EFFICIENT TRAINING OF NEURAL NETWORKS USING OPTICAL BACKPROPAGATION WITH MOMENTUM FACTOR

M.A. Otair∗ and W.A. Salameh∗∗

References

  1. [1] D.E. Rumelhart, G.E. Hinton, & R.J. Williams, Learning in-ternal representations by error propagation, in D.E. Rumel-hart & J.L. McClelland (Eds.), Parallel distributed processing(Cambridge, MA: MIT Press, 1986), 318–362.
  2. [2] M.A. Otair & W.A. Salameh, An improved back-propagationneural networks using a modified non-linear function, Proc.of the IASTED International Conf., Innsbruck, Austria, 2004,442–447.
  3. [3] F.M. Silva & L.B. Almeida, Speeding up backpropagation,in Rolf Eckmiller (Ed.), Advanced neural computers (North-Holland, 1990), 151–158.
  4. [4] R.A. Jacobs, Increased rates of convergence through learningrate adaptation, Neural Networks, 1, 1988, 169–180.
  5. [5] A.A. Minai & R.D. Williams, Acceleration of back-propagationthrough learning rate momentum adaptation, Proc. of theInternational Joint Conf. on Neural Networks, Washington,DC, USA, 1990, 1676–1679.
  6. [6] M.A. Otair & W.A. Salameh, Speeding up backpropagationneural networks, Proc. of the Informing Science and Infor-mation Technology Education Conf., Arizona, USA, 2005,167–173.
  7. [7] T. Tollenaere, SuperSAB: Fast adaptive backpropagation withgood scaling properties, Neural Networks, 3, 1990, 561–573.
  8. [8] X.H. Yu & G.A. Chen, Efficient estimation of dynamicallyoptimal learning rate and momentum for backpropagationlearning, Proc. of IEEE ICNN-95, Perth, Australia, 1995,385–388.
  9. [9] Y. Dali & L. Zemin, A LMS algorithm with stochastic momen-tum factor, Proc. of ISCAS-93, San Diego, CA, USA, 1993,1250–1253.
  10. [10] T.M. Mitchell, Machine learning (Boston, MA: McGraw Hill,1997), 119.
  11. [11] S.E. Fahlman, Faster-learning variations on backpropagation:An empirical study, Proc. of the 1988 Connectionist ModelsSummer School, (San Mateo, CA: Morgan Kaufmann, 1988),38–51.
  12. [12] M. Hagiwara, Theoretical derivation of momentum term inbackpropagation, International Joint Conf. on Neural Net-works, Baltimore, MD, USA, 1, 1992, 682–686.
  13. [13] M. Riedmiller & H.A. Braun, Direct adaptive method for fasterbackpropagation learning the PROP algorithm, Proc. of theIEEE International Conf. on Neural Networks (ICNN), I, SanFrancisco, CA, 1993, 586–591.
  14. [14] M.A. Otair & W.A. Salameh, Solving the exclusive-OR prob-lem using an optical backpropagation algorithm, Proc. of the2nd Jordan International Conf. on Computer Science andEngineering, Amman, Jordan, 2007.
  15. [15] M.A. Otair & W.A. Salameh, Enhanced version of delta-bar-delta, Proc. of the 2nd International Conf. on InformationTechnology, Amman, Jordan, 2005, 194–201.
  16. [16] M.A. Otair & W.A. Salameh, Online handwritten characterrecognition using an optical backpropagation neural networks,Proc. of the 2004 International Research Conf. on Innovationsin Information Technology, Dubai, UAE, 2004, 334–341.

Important Links:

Go Back