patrick-kidger / patrick-kidger/diffrax

Feature Request: Complex-Valued Integration With ZVODE - CVODE in Jax (autodiff)

Open
#477 5 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

question
Dominant language
Python
Stars
2.1k
Forks
189
Avg merge
3d 18h
Merged PRs (30d)
1

Description

Hi,

Unfortunately, all the available (S)ODE integration subroutines in auto-differentiable Python frameworks (RK45, Dopri, etc.) behave very poorly with complex-valued functions [*]. In the Python ecosystem, only Scipy's Fortran wrappers titled ode (ZVODE) and complex_ode (using CVODE) seem to be working fine, but obviously, they are not differentiable and not applicable to the modern applications we love.

I was wondering if anybody wants to adapt these features to Diffrax, and make them auto-differentiable.

[*] https://arxiv.org/abs/2406.06361

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named. Start by surveying Diffrax's existing ODE solver and autodiff architecture, then compare the requested complex-valued ZVODE/CVODE behavior with the current solver capabilities. Done requires a defined implementation scope, differentiable complex-valued integration, and tests demonstrating the target behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
22/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.