论文信息 - Number of hidden nodes for shape preserving ANN representation of a curve

Number of hidden nodes for shape preserving ANN representation of a curve

Scientific knowledge is often available in handbooks and journals in the form of curves. ANN representation of curves is necessary for including them in the knowledge base of a connectionist expert system. Apart from one input node and one output node, representing abscissa and ordinate of the curve, respectively, the ANN has a hidden layer with s neurons, with sigmoid activation function. A method is developed to choose the value of s, logically, and for fast determination of weights and bias values, ensuring shape preservation of the curve. The method is applied to three examples for which the results of conventional techniques like backpropagation, RBF etc. are available in literature. The comparison brings out the advantages of the method developed.

[1] Juan Manuel Peña,et al. Shape preserving representations for trigonometric polynomial curves , 1997, Comput. Aided Geom. Des..

[2] Pascal Larzabal,et al. On initialization of neural network , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[3] Stephen I. Gallant,et al. Neural network learning and expert systems , 1993 .

[4] Russell Reed,et al. Pruning algorithms-a survey , 1993, IEEE Trans. Neural Networks.

[5] M.H. Hassoun,et al. Fundamentals of Artificial Neural Networks , 1996, Proceedings of the IEEE.

[6] Kurt Hornik,et al. Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[7] B. I. Kvasov,et al. Algorithms for shape preserving local approximation with automatic selection of tension parameters , 2000, Comput. Aided Geom. Des..

[8] William C. Carpenter,et al. Common Misconceptions about Neural Networks as Approximators , 1994 .

[9] Andrew R. Webb,et al. Functional approximation by feed-forward networks: a least-squares approach to generalization , 1994, IEEE Trans. Neural Networks.

[10] Separable Regions. On Hidden Nodes for Neural Nets , 1989 .

[11] Yoshifusa Ito,et al. Approximation of continuous functions on Rd by linear combinations of shifted rotations of a sigmoid function with and without scaling , 1992, Neural Networks.

[12] Jacek M. Zurada,et al. Introduction to artificial neural systems , 1992 .