论文信息 - Improved generalization learning with sliding mode control and the Levenberg-Marquadt algorithm

Improved generalization learning with sliding mode control and the Levenberg-Marquadt algorithm

A variation of the well known Levenberg-Marquardt for training neural networks is presented in this work. The algorithm presented restricts the norm of the weights vector to a preestablished norm value and finds the minimum error solution for that norm value. A range of different norm solutions is generated and the best generalization solution is selected. The results show the efficiency of the algorithm in terms of convergence speed and generalization performance.

Antônio de Pádua Braga | Marcelo Azevedo Costa | Benjamin Rodrigues de Menezes

[1] D. Marquardt. An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[2] Mohammad Bagher Menhaj,et al. Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[3] Elie Bienenstock,et al. Neural Networks and the Bias/Variance Dilemma , 1992, Neural Computation.

[4] Mokhtar S. Bazaraa,et al. Nonlinear Programming: Theory and Algorithms , 1993 .

[5] Antônio de Pádua Braga,et al. Neural Networks Learning with Sliding Mode Control: The Sliding Mode Backpropagation Algorithm , 1999, Int. J. Neural Syst..

[6] Ricardo H. C. Takahashi,et al. Improving generalization of MLPs with multi-objective optimization , 2000, Neurocomputing.

[7] Antônio de Pádua Braga,et al. Control of generalization with a bi-objective sliding mode control algorithm , 2002, VII Brazilian Symposium on Neural Networks, 2002. SBRN 2002. Proceedings..