Estimating singular values from QR decomposition

In summary, the conversation discusses the possibility of estimating the largest and smallest singular values of a matrix A using its QR decomposition. It is mentioned that the singular values of A are the nonzero eigenvalues of A^HA or AA^H, and that the power method can be used to compute them. The speaker also suggests that if the QR decomposition of R is known, it may be possible to compute the singular values of R instead, which would be less expensive. Finally, it is concluded that since Q is orthogonal, A and R would have the same singular values, making it possible to estimate them from the QR decomposition.
  • #1
vibe3
46
1
I have a matrix for which I know its QR decomposition: [itex]A = QR[/itex]. I want to estimate the largest and smallest singular values of [itex]A[/itex] ([itex]\sigma_1[/itex] and [itex]\sigma_n[/itex]) however in my application it is too expensive to compute the full SVD of [itex]A[/itex].

Is it possible to estimate the largest/smallest singular values from the QR decomposition? The only result I've been able to find so far is
[tex]
\left| \prod_i r_{ii} \right| = \prod_i \sigma_i
[/tex]
where [itex]r_{ii}[/itex] are the diagonal entries of [itex]R[/itex]. I'm not sure if this implies that the singular values of [itex]R[/itex] are the same as the singular values of [itex]A[/itex]. If that's true, it might be possible and less expensive for my application to compute [itex]SVD(R)[/itex] rather than [itex]SVD(A)[/itex].
 
Physics news on Phys.org
  • #2
The singular values of ##A## are the nonzero eigenvalues of ## A^H A ## or ## A A^H ##. The largest eigenvalue of a matrix can usually be computed using the power method with just a few iterations. The smallest eigenvalue of a full-rank matrix is the largest eigenvalue of the inverse, so again the power method can work; if you have hte QR the inverse is cheap to compute. If not full rank ... I'm not sure what clever methods there are.

jason
 
  • #3
The problem is that my matrix [itex]A[/itex] is very ill-conditioned, so computing eigenvalues from [itex]A^T A[/itex] is unreliable (hence the reason I'm computing the QR decomposition instead). I'm looking for a numerically stable way to estimate the singular values from the QR decomposition for ill-conditioned matrices.

Suppose we have [itex]A = QR = U \Sigma V^T[/itex]. Then suppose the SVD of [itex]R[/itex] is [itex]R = U_r \Sigma_r V_r^T[/itex].

Because [itex]Q[/itex] is orthogonal, [itex]A = Q U_r \Sigma_r V_r^T[/itex] which implies [itex]U = Q U_r[/itex] and [itex]V = V_r[/itex]. This would then mean that [itex]\Sigma = \Sigma_r[/itex] and so the singular values of [itex]R[/itex] and [itex]A[/itex] are the same. Is this correct logic?
 
Last edited:
  • #4
jasonRF said:
The singular values of ##A## are the nonzero eigenvalues of ## A^H A ## or ## A A^H ##. The largest eigenvalue of a matrix can usually be computed using the power method with just a few iterations. The smallest eigenvalue of a full-rank matrix is the largest eigenvalue of the inverse, so again the power method can work; if you have hte QR the inverse is cheap to compute. If not full rank ... I'm not sure what clever methods there are.

jason
Did you meant to write that the singular values of A are the square root of the eigenvalues of [itex] A^HA [/itex] ?
 
  • Like
Likes jasonRF
  • #5
vibe3 said:
The problem is that my matrix [itex]A[/itex] is very ill-conditioned, so computing eigenvalues from [itex]A^T A[/itex] is unreliable (hence the reason I'm computing the QR decomposition instead). I'm looking for a numerically stable way to estimate the singular values from the QR decomposition for ill-conditioned matrices.

Suppose we have [itex]A = QR = U \Sigma V^T[/itex]. Then suppose the SVD of [itex]R[/itex] is [itex]R = U_r \Sigma_r V_r^T[/itex].

Because [itex]Q[/itex] is orthogonal, [itex]A = Q U_r \Sigma_r V_r^T[/itex] which implies [itex]U = Q U_r[/itex] and [itex]V = V_r[/itex]. This would then mean that [itex]\Sigma = \Sigma_r[/itex] and so the singular values of [itex]R[/itex] and [itex]A[/itex] are the same. Is this correct logic?

Yes, this is mathematically sound. A and R would have the same singular values.
 
  • #6
rs1n said:
Did you meant to write that the singular values of A are the square root of the eigenvalues of [itex] A^HA [/itex] ?
oops. yep.
 

What is QR decomposition?

QR decomposition is a matrix decomposition technique that decomposes a matrix into an orthogonal matrix and an upper triangular matrix. It is commonly used for solving linear systems of equations and for calculating the eigenvalues and eigenvectors of a matrix.

What are singular values?

Singular values are the square roots of the eigenvalues of a matrix multiplied by its transpose. They represent the scaling factors of the right singular vectors of a matrix and are important in applications such as data compression and signal processing.

Why is it important to estimate singular values from QR decomposition?

Estimating singular values from QR decomposition allows us to determine the rank of a matrix and to identify its dominant features. This information is useful for understanding the structure of a matrix and for making predictions about its behavior.

How are singular values estimated from QR decomposition?

The singular values of a matrix can be estimated by taking the absolute values of the diagonal elements of the upper triangular matrix in the QR decomposition. These values are then sorted in descending order to determine the dominant singular values of the matrix.

What are some applications of estimating singular values from QR decomposition?

Estimating singular values from QR decomposition is commonly used in data analysis, image and signal processing, and machine learning. It can also be used for solving least squares problems and for calculating the condition number of a matrix.

Similar threads

  • Linear and Abstract Algebra
Replies
1
Views
1K
  • Linear and Abstract Algebra
Replies
1
Views
2K
Replies
1
Views
2K
Replies
2
Views
2K
Replies
10
Views
2K
  • Linear and Abstract Algebra
Replies
1
Views
1K
  • Linear and Abstract Algebra
Replies
1
Views
13K
  • Calculus and Beyond Homework Help
Replies
6
Views
3K
  • Linear and Abstract Algebra
Replies
1
Views
2K
  • General Math
Replies
6
Views
3K
Back
Top