7 faster optimization for heterodyne likelihood using jax #9
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
kazewong
deleted the
7-faster-optimization-for-heterodyne-likelihood-using-jax
branch
ThibeauWouters
pushed a commit
to ThibeauWouters/jim
that referenced
this pull request
Nov 22, 2023
…-heterodyne-likelihood-using-jax 7 faster optimization for heterodyne likelihood using jax
xuyuon
added a commit
to tsunhopang/jim
that referenced
this pull request
Sep 3, 2024
…pling_improvement Extrinsic parameter sampling improvement
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Outdated suggestions cannot be applied.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.
Instead of using scipy differential evolution, this is a patch that use more jax-native optimization code.
I tried using jaxopt's optimization routine, but somehow their code always recompile the kernel no matter, including algorithms like bfgs or lbfgs. Furthermore, it is non-trivial to vmap over lbfgs to take advantage over multiple concurrent process available on GPU, so jaxopt was actually not ideal for this purpose.
flowMCnow provides a blackbox interface for optimization that uses an evolutionary algorithm to find optimum of a general function, which work quite well with the current APIL. It makes use ofvmapquite efficiently, so it can take advantage of many parallel likelihood evaluation we can provide. It also provides a pretty simple interface, so it is easy to incorporate in our workflow.This patch only show an example on how to use the API in 'GW170817_optimize.py'. We should build this into pipeline in refactoring.