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

05/02/2023
by   Michał P. Karpowicz, et al.
0

The statement in the title is not generally true, unless C and R have full rank. Then the m by r matrix C is assumed to have r independent columns (rank r). The r by n matrix R is assumed to have r independent rows (rank r). In this case the pseudoinverse C^+ is the left inverse of C, and the pseudoinverse R^+ is the right inverse of R. The simplest proof of A^+ = R^+C^+ verifies the four Penrose identities that determine the pseudoinverse A^+ of any matrix A. Our goal is a different proof of A^+ = R^+C^+, starting from first principles. We begin with the four fundamental subspaces associated with any m by n matrix A of rank r. Those are the column space and nullspace of A and A^T. The proof of A^+ = R^+C^+ then shows that this matrix acts correctly on every vector in the column space of A and on every vector in the nullspace of A^T.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/28/2018

A Sherman-Morrison-Woodbury Identity for Rank Augmenting Matrices with Application to Centering

Matrices of the form A + (V_1 + W_1)G(V_2 + W_2)^* are considered where ...
research
02/26/2020

Recommendation on a Budget: Column Space Recovery from Partially Observed Entries with Random or Active Sampling

We analyze alternating minimization for column space recovery of a parti...
research
11/23/2021

On the Column and Row Ranks of a Matrix

Every m by n matrix A with rank r has exactly r independent rows and r i...
research
05/13/2020

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

We introduce two equations expressing the inverse determinant of a full ...
research
06/11/2022

Householder Meets Student

The Householder algorithm for the QR factorization of a tall thin n x p ...
research
07/08/2022

A new perspective on the fictitious space lemma

In the present contribution we propose a new proof of the so-called fict...
research
03/09/2021

Column randomization and almost-isometric embeddings

The matrix A:ℝ^n →ℝ^m is (δ,k)-regular if for any k-sparse vector x, ...

Please sign up or login with your details

Forgot password? Click here to reset