-
arXiv:2204.07856 [pdf, ps, other]
Optimal Learning Rates for Regularized Least-Squares with a Fourier Capacity Condition
Abstract: We derive minimax adaptive rates for a new, broad class of Tikhonov-regularized learning problems in Hilbert scales under general source conditions. Our analysis does not require the regression function to be contained in the hypothesis class, and most notably does not employ the conventional \textit{a priori} assumptions on kernel eigendecay. Using the theory of interpolation, we demonstrate that… ▽ More
Submitted 15 September, 2023; v1 submitted 16 April, 2022; originally announced April 2022.
Comments: The first version of this paper contained an error -- refer to current version
-
A Trace Theorem For Sobolev Spaces On The Sierpinski Gasket
Abstract: We give a discrete characterization of the trace of a class of Sobolev spaces on the Sierpinski gasket to the bottom line. This includes the L2 domain of the Laplacian as a special case. In addition, for Sobolev spaces of low orders, including the domain of the Dirichlet form, the trace spaces are Besov spaces on the line.
Submitted 8 May, 2019; originally announced May 2019.
Comments: 18 pages, 4 figures
MSC Class: 28A80