A Review on Optimizing Radial Basis Function Neural Network using Nature Inspired Algorithm

Main Article Content

Ms. Tanvi Gupta, Supriya P Panda, S. S. Handa, M. V. Ramana Murthy

Abstract

Radial Basis Function (RBF) is a type of feed forward neural network .This function can be applied to interpolation, chaotic time-series modeling, control engineering, image restoration, data fusion etc. In RBF network, parameters of basis functions (such as width, the position and number of centers) in the nonlinear hidden layer have great influence on the performance of the network. Common RBF training algorithms cannot possibly find the global optima of nonlinear parameters in the hidden layer, and often have too many hidden units to reach certain approximation abilities, which will lead to too large a scale for the network and decline of generalization ability. Also, RBF neural network has the advantages of adaptive and self-learning ability, but it is difficult to determine the number of hidden layer neurons, and the weights learning ability from hidden layer to the output layer is low; these deficiencies easily lead to decreasing learning ability and recognition precision. Secondly, the Swarm Intelligence Algorithms are (Meta-Heuristic) development Algorithms, which attracted much attention and appeared its ability in the last ten years within many applications such as data mining, scheduling, improve the performance of artificial neural networks (ANN) and classification. So, in this paper the work of Artificial Bee Colony (ABC), Genetic algorithm(GA), Particle swarm optimization(PSO) and Bat algorithm(BA) have been reviewed, which optimized the RBF neural network in their own terms.

Article Details

How to Cite
, M. T. G. S. P. P. S. S. H. M. V. R. M. (2017). A Review on Optimizing Radial Basis Function Neural Network using Nature Inspired Algorithm. International Journal on Future Revolution in Computer Science &Amp; Communication Engineering, 3(10), 49–57. Retrieved from http://www.ijfrcsce.org/index.php/ijfrcsce/article/view/451
Section
Articles