Open AccessA Pruning Algorithm Based on Relevancy Index of Hidden Neurons Outputs
Author(s) -
Slim Abid,
Mohamed Chtourou,
Mohamed Djemel
Publication year - 2016
Publication title -
engineering, technology and applied science research/engineering, technology and applied science research
Language(s) - English
Resource type - Journals
eISSN - 2241-4487
pISSN - 1792-8036
DOI - 10.48084/etasr.704
Subject(s) - pruning , computer science , artificial neural network , index (typography) , process (computing) , artificial intelligence , algorithm , machine learning , world wide web , agronomy , biology , operating system
Choosing the training algorithm and determining the architecture of artificial neural networks are very important issues with large application. There are no general methods which permit the estimation of the adequate neural networks size. In order to achieve this goal, a pruning algorithm based on the relevancy index of hidden neurons outputs is developed in this paper. The relevancy index depends on the output amplitude of each hidden neuron and estimates his contribution on the learning process. This method is validated with an academic example and it is tested on a wind turbine modeling problem. Compared with two modified versions of Optimal Brain Surgeon (OBS) algorithm, the developed approach gives interesting results.