TuringLang / TuringLang/EllipticalSliceSampling.jl

Potential interesting ESS improvement based on mixture of Gaussians

Open
#12 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Julia
Stars
16
Forks
7
PR merge metrics
No merged PRs in 30d

Description

The paper below proposes an interesting extension of ESS that handles models with non-Gaussian prior, and models with Gaussian priors but with informative likelihoods. The authors also motivate their algorithm for parallelisable implementation. If this work well in practice, it could be an interesting gradient-free alternative to the HMC/NUTS sampler for low to mid dimensional problems.

Nishihara, R., Murray, I., & Adams, R. P. (2014). Parallel MCMC with Generalized Elliptical Slice Sampling. Journal of Machine Learning Research: JMLR, 15(61), 2087–2112.

@imurray @robertnishihara

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 by reading the cited paper on Parallel MCMC with Generalized Elliptical Slice Sampling and reviewing the existing elliptical slice sampling implementation. Clarify which algorithmic capabilities should be added, how they fit the current API, and what tests or examples would demonstrate that the improvement is complete.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.