Dimension-free tail inequalities for sums of random matrices

@article{Hsu2011DimensionfreeTI,
title={Dimension-free tail inequalities for sums of random matrices},
author={Daniel J. Hsu and Sham M. Kakade and Tong Zhang},
journal={CoRR},
year={2011},
volume={abs/1104.1672}
}

We derive exponential tail inequalities for sums of random matrices with no dependence on the explicit matrix dimensions. These are similar to the matrix versions of the Chernoff bound and Bernstein inequality except with the explicit matrix dimensions replaced by a trace quantity that can be small even when the dimension is large or infinite. Some applications to principal component analysis and approximate matrix multiplication are given to illustrate the utility of the new bounds.

The volume of convex bodies and Banach space geometry

G. Pisier

2010

A Sums of random vector outer products The following lemma is a tail inequality for smallest and largest eigenvalues of the empirical covariance matrix of subgaussian random vectors

Litvak

Advances in Neural Information Processing Systems • 2006

Inequality with applications in statistical mechanics

B. Schölkopf

Advances in Kernel Methods — Support Vector Learning • 1999