Separation Conditions, Myopic Maps, and Criteria for Uniform Approximation of Input-Output Maps
Tóm tắt
This paper considerably extends recent discrete-time (and discrete-space) results concerning the problem of obtaining criteria under which input-output maps can be uniformly approximated arbitrarily well using a certain structure consisting of a not-necessarily linear dynamic part followed by a nonlinear memoryless section that may contain sigmoids or radial basis functions, etc. In those results certain separation conditions of the kind associated with the Stone-Weierstrass theorem play a prominent role and emerge as criteria for approximation—not just sufficient conditions under which an approximation exists. Here we give corresponding results for a much larger set of maps of interest. More specifically, corresponding results are given for an important family of continuous-time and continuous space systems. As an example, it is shown that a large class of continuous-space myopic systems with continuous inputs can be uniformly approximated arbitrarily well using just a bank of shift operators followed by a nonlinear memoryless section. This directs attention to approximants for such systems of a very different type than those discussed earlier in the literature. In another example, a new result is given concerning the uniform approximation of a large class of myopic continuous-time systems with inputs that may have discontinuities.
Tài liệu tham khảo
G. Bachman and L. Narici, Functional Analysis, New York: Academic Press, 1966.
S. Boyd and L. O. Chua, “Fading Memory and the Problem of Approximating Nonlinear Operators with Volterra Series,” IEEE Transactions on Circuits and Systems, vol. 32, 1985, pp. 1150–1161.
T. Chen and H. Chen, “Approximations of Continuous Functionals by Neural Networks with Application to Dynamical Systems,” IEEE Transactions on Neural Networks, vol. 4, 1993, pp. 910–918.
T. Chen and H. Chen, “Universal Approximation to Nonlinear Operators by Neural Networks with Arbitrary Activation Functions and its Applications to Dynamic Systems,” IEEE Transactions on Neural Networks, vol. 6, 1995, pp. 911–917.
B. D. Coleman and V. J. Mizel, “Norms and Semi-groups in the Theory of Fading Memory,” Arch. Rational Mech. Anal., vol. 23, 1966, pp. 87–123.
B. D. Coleman and V. J. Mizel, “On the General Theory of Fading Memory,” Arch. Rational Mech. Anal., vol. 6, 1968, pp. 180–231.
G. Cybenko, “Approximation by Superposition of a Single Function,” Mathematics of Control, Signals and Systems, vol. 2, 1989, pp. 303–314.
Y. W. Lee, “Contributions of Norbert Wiener to Linear Theory and Nonlinear Theory in Engineering,” Selected Papers of Norbert Wiener, Siam and M.I.T. Press, 1964, pp. 17–33.
H. N. Mhaskar and C. A. Micchelli, “Approximation by Superposition of Sigmoidal and Radial Basis Functions,” Advances in Applied Mathematics, vol. 3, 1992, pp. 350–373.
J. Park and I. W. Sandberg, “Approximation and Radial-Basis Function Networks,” Neural Computation, vol. 5, 1993, pp. 305–316.
W. Rudin, Principles of Mathematical Analysis, New York: McGraw-Hill, 1976.
I. W. Sandberg, “Structure Theorems for Nonlinear Systems,” Multidimensional Systems and Signal Processing, vol. 2, 1991, pp. 267–286. (See also the Errata in vol. 3, 1992, p. 101.)
I. W. Sandberg, “Uniform Approximation and the Circle Criterion,” IEEE Transactions on Automatic Control, vol. 38, 1993, pp. 1450–1458.
I.W. Sandberg and L. Xu, “Uniform Approximation of Multidimensional Myopic Maps,” IEEE Transactions on Circuits and Systems—I, vol. 44, 1997, pp. 472–485.
I. W. Sandberg and L. Xu, “Approximation of Myopic Systems Whose Inputs Need Not be Continuous,” Multidimensional Systems and Signal Processing, vol. 9, 1998, pp. 207–225.
I.W. Sandberg, “The Mathematical Foundations of Associated Expansions for Mildly Nonlinear Systems,” IEEE Transactions on Circuits and Systems, vol. 30, 1983, pp. 441–455.
I. W. Sandberg, “Approximately-Finite Memory and Input-Output Maps,” IEEE Transactions on Circuits and Systems, vol. 39, 1992, pp. 549–556.
I.W. Sandberg, “Separation Conditions and Approximation of Discrete-Time and Discrete-Space Systems,” Journal of Circuits, Systems, and Signal Processing, vol. 17, 1998, pp. 305–320.
J. S. Shamma, “Fading Memory Systems and Robust Stability,” Automatica, vol. 29, 1993, pp. 191–200.
B. W. Stiles, I. W. Sandberg, and J. Ghosh, “Complete Memory Structures for Approximating Nonlinear Discrete-Time Mappings,” IEEE Transactions on Neural Networks, vol. 8, 1997, pp. 1397–1409.
M. H. Stone, “A Generalized Weierstrass Approximation Theorem, Studies in Modern Analysis,” in: MAA Studies in Mathematics, R. C. Buck, Ed., vol. 1, Englewood Cliffs: Prentice-Hall, pp. 30–87.
W. A. Sutherland, Introduction to Metric and Topological Spaces, Oxford: Clarendon Press, 1975.