Approximation by neural networks with weights varying on a finite set of directions

Journal of Mathematical Analysis and Applications

Volumn 389, Issue 1, 2012, Pages 72-83

  Ismailov, Vugar E ^a  

^a INSTITUTE OF PHYSICS   (Azerbaijan)

Author keywords

Activation function; Approximation; Density; MLP model; Neural network; Orbit; Path; Weight

Indexed keywords

EID: 84855858116     PISSN: 0022247X     EISSN: 10960813     Source Type: Journal    
DOI: 10.1016/j.jmaa.2011.11.037     Document Type: Article

References (25)

1. Braess D., Pinkus A. Interpolation by ridge functions. J. Approx. Theory 1993, 73:218-236.
2. Chen T., Chen H. Approximation of continuous functionals by neural networks with application to dynamic systems. IEEE Trans. Neural Networks 1993, 4:910-918.
3. Chui C.K., Li X. Approximation by ridge functions and neural networks with one hidden layer. J. Approx. Theory 1992, 70:131-141.
4. Cybenko G. Approximation by superpositions of a sigmoidal function. Math. Control Signals Systems 1989, 2:303-314.
5. DeVore R.A., Oskolkov K.I., Petrushev P.P. Approximation by feedforward neural networks. Ann. Numer. Math. 1997, 4:261-287.
6. Gripenberg G. Approximation by neural networks with a bounded number of nodes at each level. J. Approx. Theory 2003, 122:260-266.
7. Hornik K. Approximation capabilities of multilayer feedforward networks. Neural Networks 1991, 4:251-257.
8. Ismailov V.E. On the representation by linear superpositions. J. Approx. Theory 2008, 151:113-125.
9. Ismailov V.E. On the theorem of M. Golomb. Proc. Indian Acad. Sci. Math. Sci. 2009, 119(1):45-52.
10. Ito Y. Approximation of continuous functions on Rd by linear combinations of shifted rotations of a sigmoid function with and without scaling. Neural Networks 1992, 5:105-115.
11. Khavinson S.Ya. Best Approximation by Linear Superpositions (Approximate Nomography). Transl. Math. Monogr. 1997, vol. 159. American Mathematical Society, Providence, RI, 175 pp. Translated from the Russian manuscript by D. Khavinson.
12. n)?. Proc. Indian Acad. Sci. Math. Sci. 2003, 113:77-86.
13. Leshno M., Lin V.Ya., Pinkus A., Schocken S. Multilayer feedforward networks with a non-polynomial activation function can approximate any function. Neural Networks 1993, 6:861-867.
14. Maiorov V., Pinkus A. Lower bounds for approximation by MLP neural networks. Neurocomputing 1999, 25:81-91.
15. Maiorov V. Approximation by neural networks and learning theory. J. Complexity 2006, 22:102-117.
16. Makovoz Y. Uniform approximation by neural networks. J. Approx. Theory 1998, 95:215-228.
17. Marshall D.E., O'Farrell A.G. Uniform approximation by real functions. Fund. Math. 1979, 104:203-211.
18. Mhaskar H.N. On the tractability of multivariate integration and approximation by neural networks. J. Complexity 2004, 20:561-590.
19. Navada K.G. Some remarks on good sets. Proc. Indian Acad. Sci. Math. Sci. 2003, 114(4):389-397.
20. Pinkus A. Approximation theory of the MLP model in neural networks. Acta Numer. 1999, 8:143-195.
21. Pinkus A. Approximating by ridge functions. Surface Fitting and Multiresolution Methods 1997, 279-292. Vanderbilt Univ. Press, Nashville. A. Le Méhauté, C. Rabut, L.L. Schumaker (Eds.).
22. Schwartz L. Theorie generale des fonctions moyenne-periodiques. Ann. of Math. 1947, 48:857-928.
23. Sproston J.P., Strauss D. Sums of subalgebras of C(X). J. Lond. Math. Soc. 1992, 45:265-278.
24. Sternfeld Y. Uniformly separating families of functions. Israel J. Math. 1978, 29:61-91.
25. Stinchcombe M., White H. Approximating and learning unknown mappings using multilayer feedforward networks with bounded weights. Proceedings of the International Joint Conference on Neural Networks, vol. 3 1990, 7-16. IEEE, New York.