Function approximation using feedforward networks with sigmoidal signals

International Journal of Soft Computing

Volumn 1, Issue 1, 2006, Pages 76-82

  Ramakrishnan, M ^a   Velu, C M ^a   Prabakaran, N ^a   Ekambaravanan, K ^a   Vivekanandan, P ^a   Thangavelu, P ^b  

^a ANNA UNIVERSITY   (India)

^b ADITANAR COLLEGE OF ARTS AND SCIENCE   (India)

Author keywords

Feedforward networks; Function approximation; Left Sigmoidal; Right sigmoidal

Indexed keywords

EID: 70350053809     PISSN: 18169503     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (9)

1. Chen, T.H. ChenandR.W. Liu, 1995. 'Approximation Capability in C(Rn) by Multilayer Feedforward Networks and Related Problems', IEEE Transactions on Neural Network, 6: 25-29.
2. Huang, G.B. and H.A. Babri, 1998. Comments in Approximation Capability in C(R ) by Multiplayer Feedforward Networks and Related Problems'. IEEE Transactions on Neural Networks, 9: 714-715.
3. Anonymous, 1998. 'Upper Bounds on the Number of Hidden Neurons in Feedforward Networks with Arbitrary Bounded Nonlinear Activation Functions', IEEE Transactions on Neural Networks, 9: 224-229.
4. Kalmogorov, AN., 1963. On the representations of continuous functions of many variables by superposition of continuous functions of one variable, and addition. Doklady Akademy Nauk SSSR, 114(1957), 953-956.Enghsh Translation: Amer. Math. Soc. Transl, 2: 55-59.
5. Funahashi, K.I., 1989. On the Approximate Realization of Continuous Mapping by Neural Networks, Neural Networks, 2: 183-192.
6. Hornik. K., M. Stinchcombe, and H. White, 1989. Multi-layer feed forward networks are Universal approximators. Neural Networks, 2: 359-366.
7. Cybenko, G., 1989. Approximation by superpositions of a sigmoidal functions. Math. Contr. Signals. Syst, 2: 303-314.
8. Verakurkov, A., 1992. Kolmogorv's Theorem a Multilayer Neural Networks. Neural Networks, 5: 501-506.
9. Ito, Y., 1992. Approximation of continuous functions in Rd by linear combinations of shifted Rotations of a sigmoidal function with, and without scaling. Neural Networks, 5: 105-115.