CSBPRNN: A new hybridization technique using Cuckoo Search to train Back Propagation Recurrent Neural Network

Abstract

Nature inspired meta-heuristic algorithms provide derivative-free solution to optimize complex problems. Cuckoo Search (CS) algorithm is one of the most modern addition to the group of nature inspired optimization meta-heu-ristics. The Simple Recurrent Networks (SRN) were initially trained by Elman with the standard back propagation (SBP) learning algorithm which is less capable and often takes enormous amount of time to train a network of even a moderate size. And the complex error surface of the SBP makes many training algorithms are prone to being trapped in local minima. This paper proposed a new meta-heuristic based Cuckoo Search Back Propagation Recurrent Neural Network (CSBPRNN) algorithm. The CSBPRNN is based on Cuckoo Search to train BPRNN in order to achieve fast convergence rate and to avoid local minima problem. The performance of the proposed CSBPRNN is compared with Artificial Bee Colony using BP algorithm, and other hybrid variants. Specifically OR and XOR datasets are used. The simulation results show that the computational efficiency of BP training process is highly enhanced when coupled with the proposed hybrid method. © Springer Science+Business Media Singapore 2014.