stan-dev / stan-dev/stan

Enable optimization within .stan programs

Open
#1,319 4 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature
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

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

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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.