LinearEllipticalSliceSampler Robustness Improvements #1859
Closed
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
facebook-github-bot
added
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D46422349 |
Codecov Report
@@ Coverage Diff @@
## main #1859 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 172 172
Lines 15102 15111 +9
=========================================
+ Hits 15102 15111 +9
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
SebastianAment
added a commit
to SebastianAment/botorch
that referenced
this pull request
Jun 5, 2023
Summary: Pull Request resolved: pytorch#1859 This commit improves the robustness of the linear elliptical slice sampler, primarily with two modifications: 1) A rewrite of the computation of the angles of the ellipse that lead to intersections with the constraint boundaries that gets ride of the `delta_theta` parameter which cannot be set universally without sacrificing either correctness or causing errors due to floating point imprecisions. 2) Contracting the feasible slices of the ellipse by an amount close to the numerical precision to guarantee that the resulting samples satisfy the constraints numerically. The commit also introduces a high dimensional test case that enforces the monotonicity of adjacent elements of the sample vectors, which originally lead to the discovery of all the issues that are fixed by the above steps. Differential Revision: D46422349 fbshipit-source-id: 5fc474975a5d798374b3d6c7e36f6b9cf9ae6f27
SebastianAment
force-pushed
the
export-D46422349
branch
from
81060ba to
8db1ca5
Compare
|
This pull request was exported from Phabricator. Differential Revision: D46422349 |
Summary: Pull Request resolved: pytorch#1859 This commit improves the robustness of the linear elliptical slice sampler, primarily with two modifications: 1) A rewrite of the computation of the angles of the ellipse that lead to intersections with the constraint boundaries that gets rid of the `delta_theta` parameter which cannot be set universally without sacrificing either correctness or causing errors due to floating point imprecisions. 2) Contracting the feasible slices of the ellipse by an amount close to the numerical precision to guarantee that the resulting samples satisfy the constraints numerically. The commit also introduces a high dimensional test case that enforces the monotonicity of adjacent elements of the sample vectors, which originally led to the discovery of all the issues that are fixed by the above steps. Reviewed By: Balandat Differential Revision: D46422349 fbshipit-source-id: 526d090d4360ab5b073d7684cfd31495541da866
SebastianAment
force-pushed
the
export-D46422349
branch
from
8db1ca5 to
35cd148
Compare
|
This pull request was exported from Phabricator. Differential Revision: D46422349 |
SebastianAment
added a commit
to SebastianAment/botorch
that referenced
this pull request
Jun 9, 2023
Summary: Pull Request resolved: pytorch#1859 This commit improves the robustness of the linear elliptical slice sampler, primarily with two modifications: 1) A rewrite of the computation of the angles of the ellipse that lead to intersections with the constraint boundaries that gets rid of the `delta_theta` parameter which cannot be set universally without sacrificing either correctness or causing errors due to floating point imprecisions. 2) Contracting the feasible slices of the ellipse by an amount close to the numerical precision to guarantee that the resulting samples satisfy the constraints numerically. The commit also introduces a high dimensional test case that enforces the monotonicity of adjacent elements of the sample vectors, which originally led to the discovery of all the issues that are fixed by the above steps. Differential Revision: https://internalfb.com/D46422349 fbshipit-source-id: c6fcf9da5f99871a5ff4291f1f698f200ec5c4c3
SebastianAment
added a commit
to SebastianAment/botorch
that referenced
this pull request
Jun 9, 2023
Summary: Pull Request resolved: pytorch#1859 This commit improves the robustness of the linear elliptical slice sampler, primarily with two modifications: 1) A rewrite of the computation of the angles of the ellipse that lead to intersections with the constraint boundaries that gets rid of the `delta_theta` parameter which cannot be set universally without sacrificing either correctness or causing errors due to floating point imprecisions. 2) Contracting the feasible slices of the ellipse by an amount close to the numerical precision to guarantee that the resulting samples satisfy the constraints numerically. The commit also introduces a high dimensional test case that enforces the monotonicity of adjacent elements of the sample vectors, which originally led to the discovery of all the issues that are fixed by the above steps. Differential Revision: https://internalfb.com/D46422349 fbshipit-source-id: 88e85a566a963eef7403114e31c2e17f05ad1182
|
This pull request has been merged in ef52ea9. |
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 commit improves the robustness of the linear elliptical slice sampler, primarily with two modifications:
delta_thetaparameter which cannot be set universally without sacrificing either correctness or causing errors due to floating point imprecisions.The commit also introduces a high dimensional test case that enforces the monotonicity of adjacent elements of the sample vectors, which originally led to the discovery of all the issues that are fixed by the above steps.
Differential Revision: D46422349