Use float-epsilon calculation in Sabre best scores #9117
Merged
Conversation
This file contains 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
jakelishman
added
Changelog: Bugfix
|
Thank you for opening a new pull request. Before your PR can be merged it will first need to pass continuous integration tests and be reviewed. Sometimes the review process can be slow, so please be patient. While you're waiting, please feel free to review other open PRs. While only a subset of people are authorized to approve pull requests for merging, everyone is encouraged to review open pull requests. Doing reviews helps reduce the burden on the core team and helps make the project's code better for everyone. One or more of the the following people are requested to review this:
|
This should not meaningfully affect any small-scale outputs, but at
larger scales, there are floating-point instabilities that can make the
"best swaps" set miss swaps that are logically at the same score as
others. This does not (cannot) fully stabilise us against all minor
floating-point rounding errors in calculations; the order of iteration
through the swaps can still technically change the swap set, for example
if three swaps with relative differences `{0, 0.8, 1.6} * BEST_EPSILON`
are encountered in a different orders. The float epsilon is chosen to
be significantly larger than the actual magnitude of fluctuations, and
significantly smaller than the changes caused by all the actual
components of scoring, in order to mitigate this effect.
jakelishman
force-pushed
the
sabre/epsilon-compare
branch
from
b17c97d
to
8062860
Compare
Pull Request Test Coverage Report for Build 3463181611
💛 - Coveralls |
mtreinish
approved these changes
Nov 14, 2022
Cryoris
pushed a commit
to Cryoris/qiskit-terra
that referenced
this pull request
Jan 12, 2023
This should not meaningfully affect any small-scale outputs, but at
larger scales, there are floating-point instabilities that can make the
"best swaps" set miss swaps that are logically at the same score as
others. This does not (cannot) fully stabilise us against all minor
floating-point rounding errors in calculations; the order of iteration
through the swaps can still technically change the swap set, for example
if three swaps with relative differences `{0, 0.8, 1.6} * BEST_EPSILON`
are encountered in a different orders. The float epsilon is chosen to
be significantly larger than the actual magnitude of fluctuations, and
significantly smaller than the changes caused by all the actual
components of scoring, in order to mitigate this effect.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
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
This should not meaningfully affect any small-scale outputs, but at larger scales, there are floating-point instabilities that can make the "best swaps" set miss swaps that are logically at the same score as others. This does not (cannot) fully stabilise us against all minor floating-point rounding errors in calculations; the order of iteration through the swaps can still technically change the swap set, for example if three swaps with relative differences
{0, 0.8, 1.6} * BEST_EPSILONare encountered in a different orders. The float epsilon is chosen to be significantly larger than the actual magnitude of fluctuations, and significantly smaller than the changes caused by all the actual components of scoring, in order to mitigate this effect.Details and comments
This detail is (to the best of my knowledge), the only reason that #9012 broke RNG compatibility in large circuits with the previous Rust version.
Tagging "bugfix" for the more developer-facing GitHub release notes, but there's no specific note for this in the public documentation; that will come with a later PR.