Generalization Capability of Artificial Neural Network Incorporated with Pruning Method
- 2.3k Downloads
In any real world application, the performance of Artificial Neural Networks (ANN) is mostly depends upon its generalization capability. Generalization of the ANN is ability to handle unseen data. The generalization capability of the network is mostly determined by system complexity and training of the network. Poor generalization is observed when the network is over-trained or system complexity (or degree of freedom) is relatively more than the training data. A smaller network which can fit the data will have the k good generalization ability. Network parameter pruning is one of the promising methods to reduce the degree of freedom of a network and hence improve its generalization. In recent years various pruning methods have been developed and found effective in real world applications. Next, it is important to estimate the improvement in generalization and rate of improvement as pruning being incorporated in the network. A method is developed in this research to evaluate generalization capability and rate of convergence towards the generalization. Using the proposed method, experiments have been conducted to evaluate Multi-Layer Perceptron neural network with pruning being incorporated for handwritten numeral recognition.
KeywordsNeural network pruning generalization capability rate of generalization
Unable to display preview. Download preview PDF.
- 2.Lawrence, S., Giles, C.L., Tsoi, A.C.: What Size Neural Network Gives Optimal Generalization? Convergence Properties of Back propagation. Technical Report, UMIACS-TR-96-22 and CS-TR-3617, Institute for Advanced Computer Studies University of Maryland, College Park, MD 20742 (1996)Google Scholar
- 3.Lawrence, S., Giles, L.C.: Overfitting and Neural Networks: Conjugate Gradient and Backpropagation. In: IEEE IJCNN, Italy, CA, July 24-27, pp. 114–119 (2000)Google Scholar
- 5.Cun, Y.L., Denker, J.S., Solla, S.A.: Optima Brain Damage. In: Denver (ed.) Advances in Neural Information Processing (2), pp. 598–605 (1989/1990); D.S. TouretzkyGoogle Scholar
- 6.Hassibi, B., Stork, D.G.: Second-order derivatives for network pruning: Optimal Brain Surgeon. Advances in Neural Information Processing System 5, 164–171 (1993)Google Scholar
- 8.Kavzoglu, T., Mather, P.M.: Assessing Artificial Neural Network Pruning Algorithms. In: Proceedings of the 24th Annual Conference and Exhibition of the Remote Sensing Society, Greenwich, UK, pp. 603–609 (1998)Google Scholar
- 9.Kavzoglu, T., Vieria, C.A.O.: An Analysis of Artificial Neural Network Pruning Algorithms in Relation to Land Cover Classification Accuracy. In: Proceedings of the Remote Sensing Society Student Conference, Oxford, UK, pp. 53–58 (1998)Google Scholar
- 11.Galerne, P., Yao, G.K.: Burel: New Neural Network Pruning and its application to sonar Imagery. In: Conference IEEE-CESA 1998 (1998)Google Scholar
- 13.Vapnik, V.N.: Estimation of Dependencies Based on Empirical Data. Springer, Berlin (1982)Google Scholar
- 14.Bengio, Y.: Neural Networks for Speech and Sequence Recognition. Thomson (1996)Google Scholar
- 15.Urolagin, S., Prema, K.V., Reddy, N.V.S.: Applying bottom up neural engineering approach to handwritten number recognition. In: International Conference on Cognition and Recognition, India, December 22-23, pp. 469–476 (2005)Google Scholar