https://github.com/bat/mgvi.jl

Metric Gaussian Variational Inference

https://github.com/bat/mgvi.jl

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
    Links to: arxiv.org
  • Committers with academic emails
    2 of 8 committers (25.0%) from academic institutions
  • Institutional organization owner
    Organization bat has institutional domain (www.mpp.mpg.de)
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (11.4%) to scientific vocabulary

Keywords

bayesian-inference bayesian-statistics julia machine-learning posterior-distributions statistics

Keywords from Contributors

mcmc posterior prior stochastic-processes julia-package julialang raytracer pde particle numerical
Last synced: 5 months ago · JSON representation

Repository

Metric Gaussian Variational Inference

Basic Info
  • Host: GitHub
  • Owner: bat
  • License: other
  • Language: Julia
  • Default Branch: main
  • Homepage:
  • Size: 42 MB
Statistics
  • Stars: 8
  • Watchers: 7
  • Forks: 2
  • Open Issues: 2
  • Releases: 14
Topics
bayesian-inference bayesian-statistics julia machine-learning posterior-distributions statistics
Created over 5 years ago · Last pushed 6 months ago
Metadata Files
Readme License

README.md

MGVI.jl

Documentation for stable version Documentation for development version License Build Status Codecov

This is an implementation of the Metric Gaussian Variational Inference (MGVI) algorithm in julia

MGVI is an iterative method that performs a series of Gaussian approximations to the posterior. It alternates between approximating the covariance with the inverse Fisher information metric evaluated at an intermediate mean estimate and optimizing the KL-divergence for the given covariance with respect to the mean. This procedure is iterated until the uncertainty estimate is self-consistent with the mean parameter. We achieve linear scaling by avoiding to store the covariance explicitly at any time. Instead we draw samples from the approximating distribution relying on an implicit representation and numerical schemes to approximately solve linear equations. Those samples are used to approximate the KL-divergence and its gradient. The usage of natural gradient descent allows for rapid convergence. Formulating the Bayesian model in standardized coordinates makes MGVI applicable to any inference problem with continuous parameters.

Documentation

Citing MGVI.jl

When using MGVI.jl for research, teaching or similar, please cite MGVI:

@article{knollmüller2020metric, title={Metric Gaussian Variational Inference}, author={Jakob Knollmüller and Torsten A. Enßlin}, year={2020}, eprint={1901.11033}, archivePrefix={arXiv}, primaryClass={stat.ML} }

Owner

  • Name: Bayesian analysis toolkit
  • Login: bat
  • Kind: organization
  • Email: bat@mpp.mpg.de

GitHub Events

Total
  • Create event: 7
  • Commit comment event: 8
  • Release event: 4
  • Delete event: 4
  • Issue comment event: 10
  • Push event: 46
  • Pull request event: 10
Last Year
  • Create event: 7
  • Commit comment event: 8
  • Release event: 4
  • Delete event: 4
  • Issue comment event: 10
  • Push event: 46
  • Pull request event: 10

Committers

Last synced: almost 3 years ago

All Time
  • Total Commits: 94
  • Total Committers: 8
  • Avg Commits per committer: 11.75
  • Development Distribution Score (DDS): 0.521
Top Committers
Name Email Commits
Victor Ananyev v****0@g****m 45
Oliver Schulz o****z@m****e 35
github-actions[bot] 4****]@u****m 7
CompatHelper Julia c****y@j****g 3
apmypb 4****b@u****m 1
Scott Hayashi s****i@t****e 1
Jakob Knollmüller j****r@M****l 1
jknollm j****b@k****e 1
Committer Domains (Top 20 + Academic)

Issues and Pull Requests

Last synced: about 1 year ago

All Time
  • Total issues: 1
  • Total pull requests: 51
  • Average time to close issues: less than a minute
  • Average time to close pull requests: 13 days
  • Total issue authors: 1
  • Total pull request authors: 7
  • Average comments per issue: 14.0
  • Average comments per pull request: 1.1
  • Merged pull requests: 38
  • Bot issues: 0
  • Bot pull requests: 18
Past Year
  • Issues: 0
  • Pull requests: 5
  • Average time to close issues: N/A
  • Average time to close pull requests: about 1 month
  • Issue authors: 0
  • Pull request authors: 2
  • Average comments per issue: 0
  • Average comments per pull request: 1.0
  • Merged pull requests: 4
  • Bot issues: 0
  • Bot pull requests: 1
Top Authors
Issue Authors
  • JuliaTagBot (1)
Pull Request Authors
  • github-actions[bot] (18)
  • oschulz (18)
  • vindex10 (17)
  • dependabot[bot] (2)
  • apmypb (1)
  • jknollm (1)
  • sthayashi (1)
Top Labels
Issue Labels
Pull Request Labels
WIP (2) dependencies (2)

Dependencies

.github/workflows/CompatHelper.yml actions
.github/workflows/TagBot.yml actions
  • JuliaRegistries/TagBot v1 composite
.github/workflows/ci.yml actions
  • actions/cache v2 composite
  • actions/checkout v2 composite
  • codecov/codecov-action v1 composite
  • julia-actions/julia-buildpkg latest composite
  • julia-actions/julia-docdeploy latest composite
  • julia-actions/julia-processcoverage v1 composite
  • julia-actions/julia-runtest latest composite
  • julia-actions/setup-julia latest composite