The Effect of Bat Population in Bat-BP Algorithm
A new metaheuristic based back-propagation algorithm known as Bat-BP is presented in this paper. The proposed Bat-BP algorithm successfully solves the problems like slow convergence to global minima and network stagnancy in back-propagation neural network (BPNN) algorithm. In this paper, the bat population is increased from 10 to 500 bats to detect the performance decline or incline in the Bat-BP algorithm by performing simulations on XOR and OR datasets. The simulation results show that the convergence rate to global minimum in Bat-BP is directly proportional with an increase in bats on 2-bit XOR dataset. In case of 3-bit XOR and 4-bit OR datasets, the results deteriorated with an increase in the bat population.
KeywordsBat population Metaheuristics Slow convergence Global minima Bat algorithm Back-propagation neural network algorithm Bat-BP algorithm Momentum
The Authors would like to thank Office of Research, Innovation, Commercialization and Consultancy Office (ORICC), Universiti Tun Hussein Onn Malaysia (UTHM) and Ministry of Higher Education (MOHE) Malaysia for financially supporting this Research under Fundamental Research Grant Scheme (FRGS) vote no. 1236.
- 1.Deng WJ, Chen WC, Pei W (2008) Back-propagation neural network based importance-performance analysis for determining critical service attributes. J Expert Syst Appl 34(2)Google Scholar
- 2.Kosko B (1992) Neural network and fuzzy systems, 1st edn. Prentice Hall, Englewood CliffsGoogle Scholar
- 3.Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representations by error propagation. Parallel Distrib Process Explor Microstruct CognGoogle Scholar
- 4.Rehman MZ, Nawi NM, Ghazali R (2012) Studying the effect of adaptive momentum in improving the accuracy of gradient descent back propagation algorithm on classification problems. Int J Mod Phys (IJMPCS), 1(1)Google Scholar
- 5.Nawi NM, Ransing MR, Ransing RS (2007) An improved conjugate gradient based learning algorithm for back propagation neural networks. Int J Comput Intell 4(1):46–55Google Scholar
- 6.Nawi NM, Rehman MZ, Ghazali MI (2011) Noise-induced hearing loss prediction in Malaysian industrial workers using gradient descent with adaptive momentum algorithm. Int Rev Comput Softw (IRECOS), 6(5):740–749Google Scholar
- 7.Lee K, Booth D, Alam PA (2005) Comparison of supervised and unsupervised neural networks in predicting bankruptcy of Korean firms. Expert Syst Appl 29(1):1–16Google Scholar
- 8.Mendes R, Cortez P, Rocha M, Neves J (2002) Particle swarm for feed forward neural network training. In: Proceedings of the international joint conference on neural networks, vol 2, pp 1895–1899Google Scholar
- 9.Nandy S, Sarkar PP, Das A (2012) Training a feed-forward neural network with artificial bee colony based backpropagation method. Int J Comput Sci Inf Technol (IJCSIT),4(4):33–46Google Scholar
- 10.Karaboga D, Akay B, Ozturk C (2007) Artificial bee colony (ABC) optimization algorithm for training feed-forward neural networks. In: 4th international conference on modeling decisions for artificial intelligence (MDAI 2007), Kitakyushu, Japan, 16–18 Aug 2007Google Scholar
- 12.Montana DJ, Davis L (1989) Training feedforward neural networks using genetic algorithms. In: Proceedings of the eleventh international joint conference on artificial Intelligence, vol 1, pp 762–767Google Scholar
- 13.Yang XS (2010) A new metaheuristic bat-inspired algorithm. In: Nature inspired cooperative strategies for optimization (NICSO 2010), pp 65–74Google Scholar