Resumen: En este trabajo se presentan los resultados experimentales del control neuronal de velocidad de un servosistema. La estrategia de control neuronal implementada es el control por modelo inverso. El entrenamiento de la red se ha realizado usando dos algoritmos de aprendizaje: Levenberg-Marquardt y regularización Bayesiana. Se evalúa la capacidad de generalización de cada método en funció...

Abstract—Levenberg-Marquardt method (LM) was proposed to be applied as a non-linear least-square fitting in the analysis of a natural gamma-ray spectrum that was taken by the Hp (Ge) detector. The Gaussian function that composed of three components, main Gaussian, a step background function and tailing function in the lowenergy side, has been suggested to describe each of the -ray lines mathem...

In this paper, the problem of nonnegative matrix factorization (NMF) is considered. It is formulated as the optimization of a criterion with bound constraints. We propose an approach based on Givens parameterization of some positive vector, and criterion minimization is achieved using Levenberg-Marquardt algorithm. The performance of the developed NMF method is illustrated for the separation of...

Efficient second order algorithm for training feedforward neural networks is presented. The algorithm has a similar convergence rate as the Lavenberg-Marquardt (LM) method and it is less computationally intensive and requires less memory. This is especially important for large neural networks where the LM algorithm becomes impractical. Algorithm was verified with several examples.

*This research was carried out as part of NWO research project 611-304-019, 'Address: Free University, Department of Economics and Econometrics, De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands. E-mail: [email protected].