Proving the Inequality for Pseudoinverse and 2-Norm: Is ||A+|| ≤ ||A1-1||?

  • Thread starter Codezion
  • Start date
In summary, the conversation discusses how to prove the inequality ||A^+||_2 \le ||A_1^{-1}||_2, where A^+ is the pseudo-inverse of A and A_1 is a nonsingular square matrix. The attempt at a solution involved replacing A with A_1 and another random matrix in the pseudo-inverse formula, but the correct approach is to use the fact that (A^TA)^-1=A^-1(A^T)^-1. However, the problem as stated in the book is slightly different, asking to show ||A^+||_2 \le ||A_1^{-1}||_2 instead.
  • #1
Codezion
18
0

Homework Statement


Prove that the ||A+|| [tex]\leq[/tex] ||A1-1||
Where A+=(ATA)-1AT, ||.|| is the 2 norm and A is an mxn matrix

Homework Equations



A = [[tex]\stackrel{A1}{A2}[/tex]] where A1 is an nxn nonsingular square matrix and A2 is any random matrix that is (m-n)xn

The Attempt at a Solution



All I did was replace all the A's in the pseduoinverse with A1 and A2 and found the following:
||(A1TA1 + A2TA2)_1(A1 A2)|| but cannot proceed much. I really appreciate any help! Thank you.

Homework Statement


Homework Equations


The Attempt at a Solution

 
Last edited:
Physics news on Phys.org
  • #2
Well, (A^TA)^-1=A^-1(A^T)^-1
So you get that A dagger equals A^-1 by definition, have you wrriten the problem as it is in the book?
 
  • #4
You did not write the problem as specified in the book.

The book asks you to show that [tex]||A^+||_2 \le ||A_1^{-1}||_2[/tex] . This is not the same as [tex]||A^+||_2 \le ||A^{-1}||_2[/tex] . The latter does not even make sense because A can not have an inverse.
 
  • #5
Thank you DH! I have corrected my problem - I think all the latex syntax confused me.
 

FAQ: Proving the Inequality for Pseudoinverse and 2-Norm: Is ||A+|| ≤ ||A1-1||?

What is the pseudoinverse of a matrix?

The pseudoinverse of a matrix is a generalization of the inverse of a square matrix to non-square matrices. It is also known as the Moore-Penrose inverse and can be computed for any matrix, regardless of its size or invertibility.

How is the pseudoinverse calculated?

The pseudoinverse is calculated using the singular value decomposition (SVD) of a matrix. The pseudoinverse of a matrix A is given by A+ = VΣ+UT, where V and U are orthogonal matrices and Σ+ is the Moore-Penrose inverse of the diagonal matrix Σ.

What is the significance of the pseudoinverse in linear algebra?

The pseudoinverse has many applications in linear algebra, such as solving systems of linear equations, finding least-squares solutions, and computing the eigenvalues and eigenvectors of a matrix. It is also used in data analysis and machine learning algorithms.

What is the 2-norm of a matrix?

The 2-norm of a matrix is a measure of its magnitude or size. It is defined as the largest singular value of the matrix, which corresponds to the largest stretching factor when the matrix is applied to a unit vector. In other words, it is the square root of the largest eigenvalue of the matrix's Hermitian form.

How is the 2-norm related to the pseudoinverse?

The 2-norm of a matrix is closely related to its pseudoinverse. Specifically, the 2-norm of the pseudoinverse of a matrix A is equal to the reciprocal of the smallest singular value of A. This is known as the singular value inequality and is an important property of the pseudoinverse.

Similar threads

Replies
5
Views
3K
Replies
2
Views
1K
Replies
8
Views
2K
Replies
7
Views
2K
Replies
2
Views
3K
Back
Top