Optimal transport and applications
This work is from 2019 - present and advnaces both the theory and applications of optimal transport for machine learning, computational NLP, and signal processing. It is supported via grants from NSF (NSF CARERR 1553075, NSF RAISE), US Natick Army center/Tufts CABCS, and AFOSR and is presently one of my active techincal focus areas.
Zeviel Imani, Shuchin Aeron, and Taritree Wongjirad. “Score-based Diffusion Models for Generating Liquid Argon Time Projection Chamber Images”. In: Physical Review Letters D (2024). Accepted, to Appear.
Matthew Werenski, Shoaib Bin Masud, James M Murphy, and Shuchin Aeron. “On Rank EnergyStatistics via Optimal Transport: Continuity, Convergence, and Change Point Detection”. In: IEEE Transactions on Information Theory (2024). preprint:arXiv preprint arXiv:2302.07964, Accepted, to Appear.
Kevin C. Cheng, Eric L. Miller, Michael C. Hughes, and Shuchin Aeron. “Nonparametric and Regularized Dynamical Wasserstein Barycenters for Sequential Observations”. In: IEEE Transactions on Signal Processing 71(2023), pp. 3164–3178. doi: 10.1109/TSP.2023.3303616.
Shoaib Bin Masud, Matthew Werenski, James M Murphy, and Shuchin Aeron. “Multi- variate Soft Rank via Entropy-Regularized Optimal Transport: Sample Efficiency and Generative Modeling”. In: Journal of Machine Learning Research 24.160 (2023), pp. 1–65.
Marshall Mueller, Shuchin Aeron, James M. Murphy, and Abiy Tasissa. “Geo- metricallyRegularized Wasserstein Dictionary Learning”. In: Proceedings of 2nd Annual Workshop on Topology, Algebra, and Geometry in Machine Learning (TAG-ML). Vol. 221. Proceedings of Machine Learning Research. PMLR, 28 Jul 2023, pp. 384–403.
Mattthew E Werenski, Ruijie Jiang, Abiy Tasissa, Shuchin Aeron, and James M Murphy.“Measure Estimation in the Barycentric Coding Model”. In: International Conference on Machine Learning. PMLR. 2022, pp. 23781–23803.
Kevin Cheng, Shuchin Aeron, Michael C Hughes, and Eric L Miller. “Dynamical Wasserstein Barycenters for Time-series Modeling”. In: Advances in Neural Information Processing Systems. Vol. 34. 2021, pp. 27991–28003.
Kevin C Cheng, Shuchin Aeron, Michael C Hughes, Erika Hussey, and Eric L Miller. “Optimal transport based change point detection and time series segment clustering”. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE. 2020, pp. 6034–6038.
Kevin C Cheng, Eric L Miller, Michael C Hughes, and Shuchin Aeron. “On matched filtering for statistical change point detection”. In: IEEE Open Journal of Signal Processing 1 (2020), pp. 159–176.
Anoop Cherian and Shuchin Aeron. “Representation learning via adversarially- contrastive optimal transport”. In: International Conference on Machine Learning. PMLR. 2020, pp. 1820–1830.
Uroš Kalabić, Piyush Grover, and Shuchin Aeron. “Optimization-based in- centivization and control scheme for autonomous traffic”. In: 2020 IEEE Intelligent Vehicles Symposium (IV). 2020, pp. 1444–1449.doi: 10.1109/IV47402.2020.9304641.
Generative modeling using entropy regularized OT-based GoF test as loss function
Change Point Detection (CPD) using OT-based GoF test
Dynamical Wasserstein Barycenter (DWB) models for modeling time series with transitions.
Self-supervised learning: sampling hard negatives using OT