caa a22 4500 47580094X CHVBK 20180406123737.0 cr unu---uuuuu 170329e20001001xx s 000 0 eng 10.1023/A:1009698206772 doi (NATIONALLICENCE)springer-10.1023/A:1009698206772 Obtaining Fault Tolerant Multilayer Perceptrons Using an Explicit Regularization [Elektronische Daten] [Jose Bernier, J. Ortega, I. Rojas, E. Ros, A. Prieto] When the learning algorithm is applied to a MLP structure, different solutions for the weight values can be obtained if the parameters of the applied rule or the initial conditions are changed. Those solutions can present similar performance with respect to learning, but they differ in other aspects, in particular, fault tolerance against weight perturbations. In this paper, a backpropagation algorithm that maximizes fault tolerance is proposed. The algorithm presented explicitly adds a new term to the backpropagation learning rule related to the mean square error degradation in the presence of weight deviations in order to minimize this degradation. The results obtained demonstrate the efficiency of the learning rule proposed here in comparison with other algorithm. Kluwer Academic Publishers, 2000 backpropagation nationallicence regularization nationallicence multilayer perceptron nationallicence fault tolerance nationallicence mean square sensitivity nationallicence Bernier Jose Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut Ortega J. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut Rojas I. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut Ros E. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut Prieto A. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut Neural Processing Letters Kluwer Academic Publishers 12/2(2000-10-01), 107-113 1370-4621 12:2<107 2000 12 11063 https://doi.org/10.1023/A:1009698206772 text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1023/A:1009698206772 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Bernier Jose Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut NATIONALLICENCE 700 1- Ortega J. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut NATIONALLICENCE 700 1- Rojas I. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut NATIONALLICENCE 700 1- Ros E. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut NATIONALLICENCE 700 1- Prieto A. Departamento de Arquitectura y Tecnología de Computadores, Universidad de Granada, E18071-, Granada, Spain aut NATIONALLICENCE 773 0- Neural Processing Letters Kluwer Academic Publishers 12/2(2000-10-01), 107-113 1370-4621 12:2<107 2000 12 11063 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer