Weight optimization in recurrent neural networks with hybrid metaheuristic cuckoo search techniques for data classification

摘要

Recurrent neural network (RNN) has been widely used as a tool in the data classification. This network can be educatedwith gradient descent back propagation. However, traditional training algorithms have some drawbacks such as slow speed ofconvergence being not definite to find the global minimum of the error function since gradient descent may get stuck in localminima. As a solution, nature inspired metaheuristic algorithms provide derivative-free solution to optimize complex problems.This paper proposes a new metaheuristic search algorithm called Cuckoo Search (CS) based on Cuckoo bird’s behavior to trainElman recurrent network (ERN) and back propagation Elman recurrent network (BPERN) in achieving fast convergence rate andto avoid local minima problem. The proposed CSERN and CSBPERN algorithms are compared with artificial bee colony usingBP algorithm and other hybrid variants algorithms. Specifically, some selected benchmark classification problems are used. Thesimulation results show that the computational efficiency of ERN and BPERN training process is highly enhanced when coupledwith the proposed hybrid method.