Publication Type
Conference Proceeding Article
Version
publishedVersion
Publication Date
7-2022
Abstract
Optimal transport is a framework for comparing measures whereby a cost is incurred for transporting one measure to another. Recent works have aimed to improve optimal transport plans through the introduction of various forms of structure. We introduce novel order constraints into the optimal transport formulation to allow for the incorporation of structure. We define an efficient method for obtaining explainable solutions to the new formulation that scales far better than standard approaches. The theoretical properties of the method are provided. We demonstrate experimentally that order constraints improve explainability using the e-SNLI (Stanford Natural Language Inference) dataset that includes human-annotated rationales as well as on several image color transfer examples.
Discipline
Numerical Analysis and Scientific Computing | Transportation
Research Areas
Intelligent Systems and Optimization
Areas of Excellence
Digital transformation
Publication
Proceedings of the 39th International Conference on Machine Learning, Baltimore, Maryland, July 17-23
Volume
162
First Page
13313
Last Page
13333
Publisher
ML Research Press
City or Country
Baltimore
Citation
LIM, Fabian; WYNTER, Laura; and LIM, Shiau Hong.
Order constraints in optimal transport. (2022). Proceedings of the 39th International Conference on Machine Learning, Baltimore, Maryland, July 17-23. 162, 13313-13333.
Available at: https://ink.library.smu.edu.sg/sis_research/10313
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.