Characterization of a class of sigmoid functions with applications to neural networks

Neural Networks

Volumn 9, Issue 5, 1996, Pages 819-835

  Menon, Anil ^b   Mehrotra, Kishan ^a   Mohan, Chilukuri K ^b   Ranka, Sanjay ^b  

^a SYRACUSE UNIVERSITY   (United States)

^b NONE

Author keywords

Additive model; Cohen Grossberg Hopfield model; Fourier transform; Hypergeometric series; Legendre equation; Sigmoid functions

Indexed keywords

DIFFERENTIAL EQUATIONS; FEEDFORWARD NEURAL NETWORKS; FOURIER TRANSFORMS; MATHEMATICAL MODELS; TRANSFER FUNCTIONS;

ADDITIVE MODEL; COHEN-GROSSBERG-HOPFIELD MODEL; HYPERGEOMETRIC SERIES; LEGENDRE EQUATION; SIGMOID FUNCTIONS;

GRAPH THEORY;

ARTICLE; ARTIFICIAL NEURAL NETWORK; FOURIER TRANSFORMATION; PRIORITY JOURNAL;

EID: 0030198925     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/0893-6080(95)00107-7     Document Type: Article

References (40)

1. Albertini, F., Sontag, E., & Maillot, V. (1993). Uniqueness of weights for neural networks. In R. Mammone (Ed.), Artificial neural networks with applications in speech and vision. London: Chapman and Hall.
2. Babister, A.W. (1967). Transcendental functions satisfying nonhomogeneos linear differential equations. New York: Macmillan.
3. Bohn, E.V. (1963). The transform analysis of linear systems. Reading, MA: Addison-Wesley.
4. Carlitz L. The inverse of the error function. Pacific J. Math. 13(2):1963;459-470.
5. Cybenko G. Approximation by superposition of a sigmoidal function. Math. Control, Signals and Systems. 2:1989;303-314.
6. Davies, B. (1978). Integral transforms and their applications. New York: Springer Verlag.
7. Diaconis P., Shahshahani M. On nonlinear functions of linear combinations. SIAM J. Sci. Stat. Comput. 5:1984;175-191.
8. Elliot, L.D. (1993). A better activation function for artificial neural networks. Technical Report TR 93-8, Institute for Systems Research, University of Maryland, College Park, MD.
9. Erdélyi, A., Magnus, W., Oberhettinger, F., & Tricomi, F.G. (1953). Higher transcendental functions (Vol. 1). New York: McGraw-Hill.
10. Feller, W. (1965). An introduction to probability theory and its applications (Vol. II). New York: John Wiley.
11. Funahashi K. On the approximate realization of continuous mappings by neural networks. Neural Networks. 2(3):1989;183-192.
12. Gallant, A.R., & White, H. (1988). There exists a neural network that does not make avoidable mistakes. In 1EEE International Conf. on Neural Networks (Vol. 1, pp. 657-664). San Diego, CA.
13. Goodman, A.W. (1983). Univalent functions (Vol. I). New York: Mariner Publishing Co.
14. Graham, R.L., Knuth, D.E., & Patashnik, O. (1989). Concrete mathematics. Reading, MA: Addison-Wesley.
15. Grossberg S. Contour enhancement, short term memory, and constancies in reverberating neural networks. Studies Appl. Math. 52:1973;217-257.
16. Grossberg S. Nonlinear neural networks: principles, mechanisms and architectures. Neural Networks. 1:1988;17-61.
17. Hansen, E.R. (1975). A table of series and products. Englewood Cliffs, NJ: Prentice-Hall.
18. Harrington P. Sigmoid transfer functions in backpropagation neural networks. Analyt. Chem. 65(15):1993;2167-2168.
19. Hertz, J., Krogh, A., & Palmer, R.G. (1991). An introduction to the theory of neural computation (Vol. 1), Santa Fe Institute studies in the sciences of complexity. Redwood City, CA: Addison-Wesley.
20. Hopfield J.J., Tank D.W. Computing with neural circuits: A model. Science. 233:1986;625-633.
21. Hornik K., Stinchcombe M., White H. Multi-layer feedforward networks are universal approximators. Neural Networks. 2:1989;359-366.
22. Krantz, S.G., & Parks, H. . (1992). A primer of real analytic functions. Berlin: Birkhäuser Verlag.
23. Lippmann R.P. An introduction to computing with neural nets. IEEE ASSP Magazine. 4:1987;4-22.
24. Macintyre, A., & Sontag, E. (1993). Finiteness results for sigmoidal "neural" networks. In Proc. 25th Annual Symp. Theory Computing, San Diego. New York: Association for Computing Machinery (ACM).
25. Marks, R.J., & Arabshahi, P. (1994). Fourier analysis and filtering of a single hidden layer perceptron. In Int. Conf. Artificial Neural Networks (IEEE/ENNS), Sorrento, Italy.
26. McBride, R.E. (1970). Obtaining generating functions (Vol. 21). Berlin: Springer Verlag.
27. Minai A., Williams R. On the derivatives of the sigmoid. Neural Networks. 6(6):1993;845-853.
28. Minsky, M., & Papert, S.A. (1988). Perceptrons, an introduction to computational geometry. Cambridge, MA: The MIT Press.
29. Oberhettinger, F. (1973). Fourier transforms of distributions and their inverses. New York: Academic Press.
30. Pao, Y.H., & Sobajic, D.J. (1987). Metric synthesis and concept discovery with connectionist networks. In Proc. IEEE Systems, Man and Cybernetics Conf., Alexandria, VA.
31. Polya, G. (1949). Remarks on the characteristic function. In Proc. 4th Berkeley Symp. Math. Statist. & Probab. (pp. 115-123).
32. Polya, G. (1974). On the zeroes of the derivatives of a function and its analytic character. In R.P. Boas (Ed.), George Polya: collected works. (pp. 178-189). Cambridge, MA: MIT Press.
33. Rainville, E.D. (1964). Intermediate differential equations. New York: Macmillan.
34. Spanier, J., & Oldham, K. (1987). An atlas of functions. Washington, DC: Hemisphere.
35. Su, K.L. (1971). Time-domain synthesis of linear networks. Englewood Cliffs, NJ: Prentice-Hall.
36. Sussmann H.J. Uniqueness of weights for minimal feedforward nets with a given input-output map. Neural Networks. 5(4):1992;589-593.
37. White H. Learning in artificial neural networks: a statistical perspective. Neural Computation. 1:1989;425-464.
38. Whittaker, E.T., & Watson, G.N. (1927). Modern analysis (4th ed.). Cambridge: Cambridge University Press.
39. Widder, D.V. (1946). The Laplace transform. Princeton, NJ: Princeton University Press.
40. Wilf, H.S. (1989). Generatingfunctionology. New York: Academic Press.