Improving convergence speed of the neural network model using Meta heuristic algorithms for weight initialization
V Vishnu Priya, P. Natesan, K. Venu, E. Gothai · 2021
Neural network is widely used nowadays as it shows great result in solving the data classification and regression problems. The neuron is a basic unit in the artificial neural networks, which are strongly connected to each other by means of synaptic weights. Convergence speed of the network model depends on various factors such as weight initialization, learning rate, learning algorithm, batch size etc. This article is about to reduce the training time of weight initialization during training process of the neural network has been done with the help of cuckoo search optimization algorithm which is one of the mostly used Meta heuristic algorithms. The performance of network with various dataset using the proposed method for weight initialization is compared with other initialization methods. The effect of synaptic weights initialization using the heuristic algorithm has shown better convergence over other techniques.