Enable optimization within .stan programs
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 2.8k
- Forks
- 388
- Avg merge
- 2d 17h
- Merged PRs (30d)
- 15
Description
This would be similar to integrate_ode but instead of calling
theta <- integrate_ode(sho, ...);
it would be something like
theta <- solve_equations(system, ...); // calls Eigen::HybridNonLinearSolver
and / or
theta <- minimize(loss, ...); // calls LBFGS or something in Boost
We would have to make sure the derivatives due to the implicit function get put in the right place.
There is a review paper that discusses such constructs in Bayesian models at
http://onlinelibrary.wiley.com.ezproxy.cul.columbia.edu/doi/10.1002/9781118771051.ch11/pdf
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start from the proposed .stan constructs in this issue and review the referenced Eigen::HybridNonLinearSolver, LBFGS or Boost approaches and the linked review paper. Done would require a supported solve or minimize construct in Stan programs with derivatives from the implicit function placed correctly.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- compilers
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100