pith. sign in

arxiv: 1902.01911 · v4 · pith:TYRRCJ5Ynew · submitted 2019-02-05 · 🧮 math.ST · cs.LG· stat.ML· stat.TH

Uniform concentration and symmetrization for weak interactions

classification 🧮 math.ST cs.LGstat.MLstat.TH
keywords boundsuniformalgorithmsaveragecasecomplexitiesconcentrationderive
0
0 comments X
read the original abstract

The method to derive uniform bounds with Gaussian and Rademacher complexities is extended to the case where the sample average is replaced by a nonlinear statistic. Tight bounds are obtained for U-statistics, smoothened L-statistics and error functionals of l2-regularized algorithms.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.