Weight Optimization in Recurrent Neural Networks with Hybrid Metaheuristic Cuckoo Search Techniques for Data Classification

Pseudocode 2

Pseudocode of CSBPRNN algorithm.

() Initializes CS population size dimension and BPERN structure

() Load the training data

() While MSE < stopping criteria

() Pass the Cuckoo nests as weights to network

() Feed forward network runs using the weights initialized with CS

() The sensitivity of one layer is calculated from its previous one and the calculation of the sensitivity start from the last

   layer of the network and move backward using (17) and (18).

() Update weights and bias using (19) to (20)

() Calculate the error using (7)

() Minimize the error by adjusting network parameter using CS.

() Generate Cuckoo egg () by taking Levy flight from random nest.

() Abandon a fraction of the worst nest. Build new nest at new location via Levy flight to replace the old one.

() Evaluate the fitness of the nest, Chose a random nest

   If

   (a)  Then

   (b)

   (c)

   End if

() CS keeps on calculating the best possible weight at each epoch until the network is converged.

End While