Training Artificial Neural Networks by a Hybrid PSO-CS Algorithm

Jeng-Fung Chen,Ho-Nien Hsieh,Quang Do

doi:10.3390/a8020292

Jeng-Fung Chen, Ho-Nien Hsieh + Show 1 more

Open Access

https://doi.org/10.3390/a8020292

Copy DOI

Abstract

Presenting a satisfactory and efficient training algorithm for artificial neural networks (ANN) has been a challenging task in the supervised learning area. Particle swarm optimization (PSO) is one of the most widely used algorithms due to its simplicity of implementation and fast convergence speed. On the other hand, Cuckoo Search (CS) algorithm has been proven to have a good ability for finding the global optimum; however, it has a slow convergence rate. In this study, a hybrid algorithm based on PSO and CS is proposed to make use of the advantages of both PSO and CS algorithms. The proposed hybrid algorithm is employed as a new training method for feedforward neural networks (FNNs). To investigate the performance of the proposed algorithm, two benchmark problems are used and the results are compared with those obtained from FNNs trained by original PSO and CS algorithms. The experimental results show that the proposed hybrid algorithm outperforms both PSO and CS in training FNNs.

Highlights

The artificial neural network (ANN), a soft computing technique, has been successfully applied to many manufacturing and engineering areas [1]
For feedforward neural networks (FNNs) trained by PSO and CS (PSOCS), α min and α max were 0.01 and
We proposed a hybrid PSOCS algorithm based on the Particle swarm optimization (PSO) and Cuckoo Search (CS) algorithms

Summary

Introduction

The artificial neural network (ANN), a soft computing technique, has been successfully applied to many manufacturing and engineering areas [1]. Neural networks have the notable ability to derive meaning from complicated or imprecise data and can be used to extract patterns and detect trends that are too complicated to be recognized by either humans or traditional computing techniques. This means that neural networks have the ability to identify and respond to patterns that are similar but not identical to the ones with which they have been trained [2]. It is the most commonly used technique for classifying nonlinearly separable patterns [4,5] and approximating functions [6,7]

Methods

Results

Conclusion