Two equalities expressing the determinant of a matrix in terms of expectations over matrix-vector products

05/13/2020
by   Jascha Sohl-Dickstein, et al.
0

We introduce two equations expressing the inverse determinant of a full rank matrix 𝐀∈ℝ^n × n in terms of expectations over matrix-vector products. The first relationship is |det (𝐀)|^-1 = 𝔼_𝐬∼𝒮^n-1[ ‖𝐀𝐬‖^-n], where expectations are over vectors drawn uniformly on the surface of an n-dimensional radius one hypersphere. The second relationship is |det(𝐀)|^-1 = 𝔼_𝐱∼ q[ p(𝐀𝐱) / q(𝐱)], where p and q are smooth distributions, and q has full support.

READ FULL TEXT

page 1

page 2

research
12/19/2022

Matrix recovery from matrix-vector products

Can one recover a matrix efficiently from only matrix-vector products? I...
research
02/22/2021

Quantum query complexity with matrix-vector products

We study quantum algorithms that learn properties of a matrix using quer...
research
05/02/2023

The Pseudoinverse of A=CR is A^+=R^+C^+ (?)

The statement in the title is not generally true, unless C and R have fu...
research
02/10/2022

Low-Rank Approximation with 1/ε^1/3 Matrix-Vector Products

We study iterative methods based on Krylov subspaces for low-rank approx...
research
09/10/2023

Hutchinson's Estimator is Bad at Kronecker-Trace-Estimation

We study the problem of estimating the trace of a matrix 𝐀 that can only...
research
09/03/2023

(min,+) Matrix and Vector Products for Inputs Decomposable into Few Monotone Subsequences

We study the time complexity of computing the (min,+) matrix product o...
research
01/27/2021

Hadamard Powers and the Identification of Mixtures of Products

The Hadamard Power of a matrix is the matrix consisting of all Hadamard ...

Please sign up or login with your details

Forgot password? Click here to reset