Abstract:Randomized trace estimation is a popular and well studied technique that approximates the trace of a large-scale matrix $B$ by computing the average of $x^T Bx$ for many samples of a random vector $X$. Often, $B$ is symmetric positive definite (SPD) but a number of applications give rise to indefinite $B$. Most notably, this is the case for log-determinant estimation, a task that features prominently in statistical learning, for instance in maximum likelihood estimation for Gaussian process regression. The analysis of randomized trace estimates, including tail bounds, has mostly focused on the SPD case. In this work, we derive new tail bounds for randomized trace estimates applied to indefinite $B$ with Rademacher or Gaussian random vectors. These bounds significantly improve existing results for indefinite $B$, reducing the the number of required samples by a factor $n$ or even more, where $n$ is the size of $B$. Even for an SPD matrix, our work improves an existing result by Roosta-Khorasani and Ascher for Rademacher vectors. This work also analyzes the combination of randomized trace estimates with the Lanczos method for approximating the trace of $f(A)$. Particular attention is paid to the matrix logarithm, which is needed for log-determinant estimation. We improve and extend an existing result, to not only cover Rademacher but also Gaussian random vectors.

Block determinants, partial determinants and the exponential map

Preserving the trace of the Kronecker sum

Extensions of some matrix inequalities related to trace and partial traces

Asymptotics of block Toeplitz determinants with piecewise continuous symbols

On Some Determinants and Permanents

On certain determinants and the square root of some determinants involving Legendre Symbols

Trace and Determinant Kernels between Matrices

Inequalities regarding partial trace and partial determinant

Noncommutative Residues, Equivariant Traces, and Trace Expansions for an Operator Algebra on $\mathbb R^n$

Residence Time Distribution for a Class of Gaussian Markov Processes

Bordered and Framed Toeplitz and Hankel Determinants with Applications to Integrable Probability

On Some Determinants with Legendre Symbol Entries.

A Determinantal Identity

The g-Drazin Invertibility of a Block Operator Matrix

Determinantal Representations and the Image of the Principal Minor Map

THE GEOMETRY OF DETERMINANTS

Multislant matrices and Jacobi--Trudi determinants over finite fields

On randomized trace estimates for indefinite matrices with an application to determinants

Block Maps and Fourier Analysis

Application of quasideterminants to the inverse of block triangular matrices over noncommutative rings