Kernel interpolation in Sobolev spaces is not consistent in low dimensions

Simon Buchholz
Proceedings of Thirty Fifth Conference on Learning Theory, PMLR 178:3410-3440, 2022.

Abstract

We consider kernel ridgeless ridge regression with kernels whose associated RKHS is a Sobolev space $H^s$. We show for $d/2

Cite this Paper


BibTeX
@InProceedings{pmlr-v178-buchholz22a, title = {Kernel interpolation in Sobolev spaces is not consistent in low dimensions}, author = {Buchholz, Simon}, booktitle = {Proceedings of Thirty Fifth Conference on Learning Theory}, pages = {3410--3440}, year = {2022}, editor = {Loh, Po-Ling and Raginsky, Maxim}, volume = {178}, series = {Proceedings of Machine Learning Research}, month = {02--05 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v178/buchholz22a/buchholz22a.pdf}, url = {https://proceedings.mlr.press/v178/buchholz22a.html}, abstract = {We consider kernel ridgeless ridge regression with kernels whose associated RKHS is a Sobolev space $H^s$. We show for $d/2
Endnote
%0 Conference Paper %T Kernel interpolation in Sobolev spaces is not consistent in low dimensions %A Simon Buchholz %B Proceedings of Thirty Fifth Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2022 %E Po-Ling Loh %E Maxim Raginsky %F pmlr-v178-buchholz22a %I PMLR %P 3410--3440 %U https://proceedings.mlr.press/v178/buchholz22a.html %V 178 %X We consider kernel ridgeless ridge regression with kernels whose associated RKHS is a Sobolev space $H^s$. We show for $d/2
APA
Buchholz, S.. (2022). Kernel interpolation in Sobolev spaces is not consistent in low dimensions. Proceedings of Thirty Fifth Conference on Learning Theory, in Proceedings of Machine Learning Research 178:3410-3440 Available from https://proceedings.mlr.press/v178/buchholz22a.html.

Related Material