A package to measure the dependence between two random elements given i.i.d. pairs of them with the entropy-regularized optimal transport independence criterion (ETIC), introduced in this paper in AISTATS 2022 (Oral presentation).

This package is based on PyTorch. Other dependencies can be found in the file environment.yml. If using conda, run the following command to install all required packages and activate the environment:

$ conda env create --file environment.yml
$ conda activate etic

Clone the repository here:

$ git clone https://github.com/langliu95/etic.git
$ cd etic/

If you find this package useful, or you use it in your research, please cite:

@inproceedings{lph,
title = {Entropy Regularized Optimal Transport Independence Criterion},
author = {Liu, Lang and
          Pal, Soumik and
          Harchaoui, Zaid},
booktitle = {AISTATS},
year = {2022}
}

This work was supported by NSF CCF-2019844, NSF DMS-2023166, NSF DMS-2052239, a PIMS CRG (PIHOT), NSF DMS-2134012, CIFAR-LMB, and faculty research awards.