https://github.com/zgbkdlm/gfk
Generative diffusion posterior sampling for informative likelihoods
https://github.com/zgbkdlm/gfk
diffusion-models generative-model guided-diffusion posterior-sampling
Last synced: about 1 month ago
JSON representation
Generative diffusion posterior sampling for informative likelihoods
- Host: GitHub
- URL: https://github.com/zgbkdlm/gfk
- Owner: zgbkdlm
- License: gpl-3.0
- Created: 2024-10-28T10:10:32.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-06-03T06:05:15.000Z (about 1 year ago)
- Last Synced: 2025-06-03T17:59:50.123Z (about 1 year ago)
- Topics: diffusion-models, generative-model, guided-diffusion, posterior-sampling
- Language: Python
- Homepage:
- Size: 375 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Generative diffusion posterior sampling for informative likelihoods
This implementation is associated with the paper "Generative diffusion posterior sampling for informative likelihoods" http://arxiv.org/abs/2506.01083.
In the paper we develop a new approach for conditional sampling of generative diffusion models with sequential Monte Carlo methods.
# Installation
Install the package via a standard procedure:
```bash
git clone git@github.com:zgbkdlm/gfk.git
cd gfk
pip install -e .
```
Depending on whether you need to run in a CPU/GPU, you may want to uninstall `jax`and `jaxlib` and then reinstall.
# Reproduce experiments
To exactly reproduce the numbers and figures in the paper, first run experiments:
```bash
cd experiments
python runs_gms/bash_aux.sh --dx=256 --nparticles=16384
python runs_gms/bash_aux_noiseless.sh --dx=256 --nparticles=16384
python runs_gms/bash_mcgdiff.sh --dx=256 --nparticles=16384
python runs_gms/bash_wu.sh --dx=256 --nparticles=16384
```
Then, run the scripts in `./summary` to produce the tables and figures, e.g.,
```bash
cd experiements
python ./summary/tabulate_gms.py
```
will produce the table.
# Citation
```bibtex
@article{Zhao2025b0smc,
author = {Zhao, Zheng},
title = {Generative diffusion posterior sampling for informative likelihoods},
journal = {Communications in Information and Systems},
note = {Special issue for celebrating Thomas Kailath's 90th birthday},
year = {2025},
}
```
# Contact
Zheng Zhao, Linköping University, https://zz.zabemon.com.