Skip to content

WenboGong/MetaSGMCMC

master
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 
 
 
 
 

Meta-Learning for Stochastic Gradient MCMC

This repo contains the implementation of the experiments in the following paper:

Wenbo Gong*, Yingzhen Li* and Jose Miguel Hernandez Lobato

Meta-Learning for Stochastic Gradient MCMC

International Conference on Learning Representations (ICLR), 2019

Contributions: Wenbo and Yingzhen initiated the project together. Wenbo designed thewith architecture of the sampler, and implemented all the experiments. Yingzhen is mainly responsible for loss function design, experiment design, and the paper writing. Miguel is here because he is Wenbo's PhD supervisor, and he provided some comments on the draft.

If you use our code in your research, please consider citing the paper.

Gaussian experiments

See README in Toy Example/

MNIST experiments

See README in BNN_MNIST/

Citing the paper (bib)

@inproceedings{
gong2018metalearning,
title={Meta-Learning For Stochastic Gradient {MCMC}},
author={Wenbo Gong and Yingzhen Li and José Miguel Hernández-Lobato},
booktitle={International Conference on Learning Representations},
year={2019},
url={https://openreview.net/forum?id=HkeoOo09YX},
}

About

The code for Meta Learning for SGMCMC

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published