Analytic Method for Evaluation of the Weights of a Robust Large-Scale Multilayer Neural Network
2015 (English)In: International Journal On Advances in Networks and Services, ISSN 1942-2644, E-ISSN 1942-2644, Vol. 8, no 3-4, 139-148 p.Article in journal (Refereed) Published
The multilayer feedforward neural network is presently one of the most popular computational methods in computer science. However, the current method for the evaluation of its weights is performed by a relatively slow iterative method known as backpropagation. According to previous research on a large-scale neural network with many hidden nodes, attempts to use an analytic method for the evaluation of the weights by the linear least square method showed to accelerate the evaluation process significantly. Nevertheless, the evaluated network showed in preliminary tests to fail in robustness compared to well-trained networks by backpropagation, thus resembling overtrained networks. This paper presents the design and verification of a new method that solves the robustness issues for such a neural network, along with MATLAB code for the verification of key experiments.
Place, publisher, year, edition, pages
2015. Vol. 8, no 3-4, 139-148 p.
Analytic, big data, FNN, large-scale, least square method, multilayer, neural network, robust, sigmoid.
Research subject Computer Science; Mathematics with specialization in Applied Mathematics
IdentifiersURN: urn:nbn:se:uu:diva-270128OAI: oai:DiVA.org:uu-270128DiVA: diva2:890185
In press2015-12-312015-12-212016-07-20Bibliographically approved