Specify convention for estimate_objective
#215
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
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Contributor
|
AdvancedVI.jl documentation for PR #215 is available at: |
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Benchmark Results
| Benchmark suite | Current: b14bb9f | Previous: 51a0b3f | Ratio |
|---|---|---|---|
normal/RepGradELBO + STL/meanfield/Zygote |
3920438191 ns |
4042091118.5 ns |
0.97 |
normal/RepGradELBO + STL/meanfield/ReverseDiff |
1137314480 ns |
1148698006 ns |
0.99 |
normal/RepGradELBO + STL/meanfield/Mooncake |
1248798164 ns |
1252557751 ns |
1.00 |
normal/RepGradELBO + STL/fullrank/Zygote |
3888598013 ns |
3995457850.5 ns |
0.97 |
normal/RepGradELBO + STL/fullrank/ReverseDiff |
1660774838 ns |
1679345707 ns |
0.99 |
normal/RepGradELBO + STL/fullrank/Mooncake |
1250085352 ns |
1280789300.5 ns |
0.98 |
normal/RepGradELBO/meanfield/Zygote |
2775544305.5 ns |
2838937181.5 ns |
0.98 |
normal/RepGradELBO/meanfield/ReverseDiff |
780909736 ns |
793764611 ns |
0.98 |
normal/RepGradELBO/meanfield/Mooncake |
1086150808 ns |
1119822194 ns |
0.97 |
normal/RepGradELBO/fullrank/Zygote |
2770044101 ns |
2877874051 ns |
0.96 |
normal/RepGradELBO/fullrank/ReverseDiff |
973336667 ns |
978741132 ns |
0.99 |
normal/RepGradELBO/fullrank/Mooncake |
1113137202 ns |
1154813432 ns |
0.96 |
normal + bijector/RepGradELBO + STL/meanfield/Zygote |
5583939500 ns |
5705945858 ns |
0.98 |
normal + bijector/RepGradELBO + STL/meanfield/ReverseDiff |
2408835091 ns |
2387917982 ns |
1.01 |
normal + bijector/RepGradELBO + STL/meanfield/Mooncake |
4158213745.5 ns |
4145579318 ns |
1.00 |
normal + bijector/RepGradELBO + STL/fullrank/Zygote |
5577515624 ns |
5711272499 ns |
0.98 |
normal + bijector/RepGradELBO + STL/fullrank/ReverseDiff |
3063136975 ns |
3111756066.5 ns |
0.98 |
normal + bijector/RepGradELBO + STL/fullrank/Mooncake |
4187254056 ns |
4302876428 ns |
0.97 |
normal + bijector/RepGradELBO/meanfield/Zygote |
4312418628 ns |
4394508316 ns |
0.98 |
normal + bijector/RepGradELBO/meanfield/ReverseDiff |
2049108141 ns |
2027700486 ns |
1.01 |
normal + bijector/RepGradELBO/meanfield/Mooncake |
3927457037.5 ns |
4040389415.5 ns |
0.97 |
normal + bijector/RepGradELBO/fullrank/Zygote |
4322323571 ns |
4492494518.5 ns |
0.96 |
normal + bijector/RepGradELBO/fullrank/ReverseDiff |
2294744589 ns |
2314373240 ns |
0.99 |
normal + bijector/RepGradELBO/fullrank/Mooncake |
4054198058 ns |
4147670105 ns |
0.98 |
This comment was automatically generated by workflow using github-action-benchmark.
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.
The objectives used in variational inference are sometimes minimized or maximized, which can be confusing. For instance the KL divergence is supposed to minimized, but the ELBO is usually maximized. Allowing this to vary on an algorithm-to-algorithm basis doesn't make a lot of sense. Therefore, this PR enforces the convention that all objective values (that are only accessible by calling
estimate_objective) are expected to be minimized. For instance, for ELBO maximization algorithms,estimate_objectivewill return the negative ELBO.