The main scripts used to generate the results reported in the paper are listed below:
- Learning and Sampling on Data Samples: produces the results in Experiments on the 5×5 lattice (learning from data samples).
- Learning and Sampling on True Model: produces the results in Experiments on the 5×5 lattice (learning from the true distribution).
- Learning and Sampling on Gibbs Samples: reproduces the results in Experiments on the 10×10 lattice.
- Learning and Sampling for D-Wave: produces the results for the D-Wave dataset experiments.
All other files contain auxiliary functions and utilities that are called by these main scripts.
License: This code is provided under a BSD license as part of the Optimization, Inference and Learning for Advanced Networks project, C18014.
Reference The code presented here is part of the following work:
@article{biswal2026selecting,
title={Selecting Optimal Variable Order in Autoregressive Ising Models},
author={Biswal, Shiba and Vuffray, Marc and Lokhov, Andrey Y},
journal={arXiv preprint arXiv:2602.20394},
year={2026}
}