Improved UODE inference of Lotka Volterra#155
Merged
Merged
Conversation
Collaborator
|
I changed the base branch to #mh-uode and requested @markusheinonen to review -- that makes the most sense to me from a workflow perspective. Then afterwards we can work on merging mh-uode to main. |
DanWaxman
added a commit
that referenced
this pull request
Jun 3, 2026
* Add UODE Lotka-Volterra notebook: observe both species, tighter sparsity - Observe both prey and predator (H=I_2) at dt=0.2 (150 points) - Laplace(0, 0.005) prior for SINDy-like sparsity on Theta - SVI config: lr=1e-3, cov_rescaling=2.0, lr_decay=0.5, 2000 steps - Correctly recovers xy interaction coefficients (Frobenius err ~0.077) - Includes debug CLI script for systematic experiments * Improved UODE inference of Lotka Volterra (#155) * add MLL-truth computation and more visualizations * working lv UODE---use EnKF + 2000 adam epochs with lr=1e-3 + fixed posterior predictive filtering plots * reverted to original timeseries length * Upgrade to predict times API * Remove debug script * Fix some typos and formatting * Link UODE notebook * fix text to match code --------- Co-authored-by: Matt Levine <mattlevine22@gmail.com> Co-authored-by: Dan Waxman <dan.waxman1@gmail.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Improving #153 for UODE inference of Lotka-Volterra.
Made the following changes:
ContinuousTimeEnKF) instead of Extended, as it is more numerically robustAutoDeltaguide,svi_result.paramsdoesn't always do what you think it does.Additional thoughts:
Discretizer+Simulatorw/DiracIdentityObservationpattern, as in https://github.com/BasisResearch/dynestyx/blob/1c02f609350a98d11f3212485b8ec9601bfbe02b/docs/deep_dives/l63_speedup_dirac_vs_enkf.ipynbN.B. I took these settings (EnKF + AutoDelta with lr=1e-3) from the SINDY-notebook for Fitz-Hugh-Nagumo. It is reassuring that they generalized nicely to this case.
N.B.B. most of the 4000 epochs are to "hyper-refine" the inference; the data are explained quite well after 1000 epochs, and the final 3000 epochs mostly focus on full sparsification (to see this, rerun with 1000 epochs).