This archive is distributed in association with the INFORMS Journal on Computing under the MIT License.
The software and data in this repository are a snapshot of data and results that were used in the research reported on in the paper Inference in Higher-order Undirected Graphical Models and Binary Polynomial Optimization by Aida Khajavirad and Yakun Wang.
Important: This code is being developed on an on-going basis at https://github.com/Yakun1125/Inference-in-Higher-order-Graphical-Model. Please go there if you would like to get a more recent version or would like support.
To cite the contents of this repository, please cite both the paper and this repo, using their respective DOIs.
https://doi.org/10.1287/ijoc.2024.0776
https://doi.org/10.1287/ijoc.2024.0776.cd
Below is the BibTex for citing this snapshot of the repository.
@misc{Wang2024,
author = {Khajavirad, Aida and Wang, Yakun},
publisher = {INFORMS Journal on Computing},
title = {Inference in higher-order undirected graphical models and binary polynomial optimization},
year = {2024},
doi = {10.1287/ijoc.2024.0776.cd},
url = {https://github.com/INFORMSJoC/2024.0776},
note = {Available for download at \url{https://github.com/INFORMSJoC/2024.0776}},
}
This project explores the use of Linear Programming (LP) relaxations for inference in higher-order undirected graphical models. It specifically focuses on two practical applications:
- Image Denoising: Utilizing graphical models to recover clean image from noisy observation.
- Error-Correcting Decoding: Recover ground turth message transmitted from noisy channel.
- denoising/: Contains datasets and scripts related to the image denoising experiments.
- decoding/: Includes data and scripts for experiments on error-correcting decoding.
See the corresponding subfolders for instructions and scripts to run experiments for each application.