SUN Debao, GAO Chao. A USEFUL BP ALGORITHM TO OVERCOME LOCAL MINIMA[J]. INFORMATION AND CONTROL, 1995, 24(5): 283-287.
Citation: SUN Debao, GAO Chao. A USEFUL BP ALGORITHM TO OVERCOME LOCAL MINIMA[J]. INFORMATION AND CONTROL, 1995, 24(5): 283-287.

A USEFUL BP ALGORITHM TO OVERCOME LOCAL MINIMA

More Information
  • Received Date: June 20, 1993
  • Published Date: October 19, 1995
  • BP algorlthm is actually a nonlinear optimal problem and produces inevitably Iocal minima.The paper proposes the rules to judge local minima and introduces momentum function in BP algorithm. Thealgorithm may iump local minima,Simulation result is given to show the effectiveness and the usage of the algorithm.
  • [1]
    Rumelhart D E. McClelland J L. Parallel Distributed Processing. MIT Press. 1986:1.2.
    [2]
    Minsky M Papert S. Perceptrons. MIT Press, 1969.
    [3]
    Gori M.Tesi A.Some Examples of Local Minima During Learning with BP.Proc Parallel Archrtecture Veural Net-works,1990.
    [4]
    Werbs D J.Back-Propagation: Past and Future. 1988 IEEE International Conference on Veural Networks, 1988. 1.
    [5]
    Hush D R.Salas J M. Improving the Learning Rate of BP with the Gradient Reuse Algorithm. 1988 IEEE International Conference on Veural Vetworks.1988. 1.
    [6]
    Wessels L F A.Barnarcd E. Avoiding False Local Minima by Proper Initialirstion of Connections. IEEE Veural Networks.1992.3(6).
    [7]
    Yu X H. Can Back-Propagation Error Surface Vot Have Local Minima. IEEE Veural Vetworks.1992.3(6).
    [8]
    Baldi O. Neural Networks and Principal Component Analysis:Learning from Exmaples and Local Minima. Veural Vetworks,1989.2:53-58.
    [9]
    Gori U. Tesi A. Un the Problem of Local Minima in BP. IEEE Trans Pattern Anal Mach Intell,JAI. l992.

Catalog

    Article views (893) PDF downloads (144) Cited by()
    Related

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return