Applying a slight perturbation for ill-conditioned matrices #13882
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
One or more of the following people are relevant to this code:
|
… adding bugfix release note
jakelishman
reviewed
Feb 19, 2025
Comment on lines
251
to
252
| if apply_perturbation: | ||
| values = [v - 1e-10 for v in values] |
There was a problem hiding this comment.
Possibly we want to do values = values - 1e-10 to use the Numpy vectorised subtraction, and to keep values as an ndarray?
There was a problem hiding this comment.
Indeed! Done in 59a7b3c.
jakelishman
approved these changes
Feb 19, 2025
|
@Mergifyio backport main |
✅ Backports have been created
|
|
AArch64 passed its tests, so I'll merge. |
mergify bot
pushed a commit
that referenced
this pull request
Feb 19, 2025
* Applying a slight perturbation for ill-conditioned matrices * Further improving the fix based on discussion with Jake and Gadi, and adding bugfix release note * Jake's suggestion from code review (cherry picked from commit 69eb2aa)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
Fix two failing tests for the 1.3.3 release:
test.python.quantum_info.operators.channel.test_kraus.TestKraus.test_circuit_init+ one other very similar test.The problem only occurs on aarch64 (arm linux), however the relevant matrix is ill-conditioned on my mac as well: the choi matrix has many eigenvalues that are negative and close to 0 (the "negative" is likely due to numerical imprecision errors), with the
numpy.linalg.condreturningnp.inf.By doing a search on the internet, one of the recommended techniques is to apply regularization, in this case adding a small perturbation to the matrix. Update: as per @jakelishman's suggestion, we can in fact perfectly recover the eigenvalues of the original matrix from the matrix with the perturbation.