Modulation Spaces and the Curse of Dimensionality
R. Parhi, M. Unser
Proceedings of the Fourteenth International Workshop on Sampling Theory and Applications (SampTA'23), Yale NH, USA, July 10-14, 2023.
We investigate the L2-error of approximating functions in the modulation spaces Ms1,1(ℝd), s ≥ 0, by linear combinations of Wilson bases elements. We analyze a nonlinear method for approximating functions in Ms1,1(ℝd) with N-terms from a Wilson basis. Its L2-approximation error decays at a rate of N− ½ − s ∕ 2d. We show that this rate is optimal by proving a matching lower bound. Remarkably, these rates do not grow with the input dimension d. Finally, we show that the best linear L2-approximation error cannot decay faster than N−s ∕ 2d. This shows that linear methods, contrary to the nonlinear ones, necessarily suffer the curse of dimensionality in these spaces.
@INPROCEEDINGS(http://bigwww.epfl.ch/publications/parhi2302.html, AUTHOR="Parhi, R. and Unser, M.", TITLE="Modulation Spaces and the Curse of Dimensionality", BOOKTITLE="Proceedings of the Fourteenth International Workshop on Sampling Theory and Applications ({SampTA'23})", YEAR="2023", editor="", volume="", series="", pages="", address="Yale NH, USA", month="July 10-14,", organization="", publisher="", note="")