https://github.com/futurecomputing4ai/easyrejectionsampling
https://github.com/futurecomputing4ai/easyrejectionsampling
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/futurecomputing4ai/easyrejectionsampling
- Owner: FutureComputing4AI
- Created: 2023-07-31T01:17:56.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2023-08-01T23:25:58.000Z (almost 2 years ago)
- Last Synced: 2025-01-22T10:29:36.230Z (5 months ago)
- Language: Jupyter Notebook
- Size: 13.7 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Easy Rejection Sampling
This repository hosts our research code for the Easy Rejection Sampling (ERS) algorithm. The goal is that ERS be easy from a user's perspective - in that they only need to pass a function that computes the (unnormalized) density they wish to draw samples from, and optionally the domain of that function if not infinite. In this code any function defined in a JAX compatible way should work.
This is research quality code with no warranty or support.
## Citations
If you use the ERS algorithm or code, please cite our work!
```
@inproceedings{ERS,
author = {Raff, Edward and McLean, Mark and Holt, James},
booktitle = {European Conference on Artificial Intelligence (ECAI)},
title = {{An Easy Rejection Sampling Baseline via Gradient Refined Proposals}},
year = {2023}
}
```