Home > Research > Publications & Outputs > Constrained Hellinger–Kantorovich Barycenters

Electronic data

  • CHKB_accepted

    Accepted author manuscript, 627 KB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

Constrained Hellinger–Kantorovich Barycenters: Least-Cost Soft and Conic Multimarginal Formulations

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Constrained Hellinger–Kantorovich Barycenters: Least-Cost Soft and Conic Multimarginal Formulations. / Buze, Maciej.
In: SIAM Journal on Mathematical Analysis, Vol. 57, No. 1, 28.02.2025, p. 495-519.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Buze M. Constrained Hellinger–Kantorovich Barycenters: Least-Cost Soft and Conic Multimarginal Formulations. SIAM Journal on Mathematical Analysis. 2025 Feb 28;57(1):495-519. Epub 2025 Jan 13. doi: 10.1137/24M1639804

Author

Buze, Maciej. / Constrained Hellinger–Kantorovich Barycenters : Least-Cost Soft and Conic Multimarginal Formulations. In: SIAM Journal on Mathematical Analysis. 2025 ; Vol. 57, No. 1. pp. 495-519.

Bibtex

@article{52ddc3c813b14b8c88688483aabbacee,
title = "Constrained Hellinger–Kantorovich Barycenters: Least-Cost Soft and Conic Multimarginal Formulations",
abstract = "We show that the problem of finding the barycenter in the Hellinger–Kantorovich setting admits a least-cost soft multimarginal formulation, provided that a one-sided hard marginal constraint is introduced. The constrained approach is then shown to admit a conic multimarginal reformulation based on defining a single joint multimarginal perspective cost function in the conic multimarginal formulation, as opposed to separate two-marginal perspective cost functions for each two-marginal problem in the coupled-two-marginal formulation, as was studied previously in the literature. We further establish that, as in the Wasserstein metric, the recently introduced framework of unbalanced multimarginal optimal transport can be reformulated using the notion of the least cost. Subsequently, we discuss an example when input measures are Dirac masses and numerically solve an example for Gaussian measures. Finally, we also explore why the constrained approach can be seen as a natural extension of a Wasserstein space barycenter to the unbalanced setting.",
author = "Maciej Buze",
year = "2025",
month = feb,
day = "28",
doi = "10.1137/24M1639804",
language = "English",
volume = "57",
pages = "495--519",
journal = "SIAM Journal on Mathematical Analysis",
number = "1",

}

RIS

TY - JOUR

T1 - Constrained Hellinger–Kantorovich Barycenters

T2 - Least-Cost Soft and Conic Multimarginal Formulations

AU - Buze, Maciej

PY - 2025/2/28

Y1 - 2025/2/28

N2 - We show that the problem of finding the barycenter in the Hellinger–Kantorovich setting admits a least-cost soft multimarginal formulation, provided that a one-sided hard marginal constraint is introduced. The constrained approach is then shown to admit a conic multimarginal reformulation based on defining a single joint multimarginal perspective cost function in the conic multimarginal formulation, as opposed to separate two-marginal perspective cost functions for each two-marginal problem in the coupled-two-marginal formulation, as was studied previously in the literature. We further establish that, as in the Wasserstein metric, the recently introduced framework of unbalanced multimarginal optimal transport can be reformulated using the notion of the least cost. Subsequently, we discuss an example when input measures are Dirac masses and numerically solve an example for Gaussian measures. Finally, we also explore why the constrained approach can be seen as a natural extension of a Wasserstein space barycenter to the unbalanced setting.

AB - We show that the problem of finding the barycenter in the Hellinger–Kantorovich setting admits a least-cost soft multimarginal formulation, provided that a one-sided hard marginal constraint is introduced. The constrained approach is then shown to admit a conic multimarginal reformulation based on defining a single joint multimarginal perspective cost function in the conic multimarginal formulation, as opposed to separate two-marginal perspective cost functions for each two-marginal problem in the coupled-two-marginal formulation, as was studied previously in the literature. We further establish that, as in the Wasserstein metric, the recently introduced framework of unbalanced multimarginal optimal transport can be reformulated using the notion of the least cost. Subsequently, we discuss an example when input measures are Dirac masses and numerically solve an example for Gaussian measures. Finally, we also explore why the constrained approach can be seen as a natural extension of a Wasserstein space barycenter to the unbalanced setting.

U2 - 10.1137/24M1639804

DO - 10.1137/24M1639804

M3 - Journal article

VL - 57

SP - 495

EP - 519

JO - SIAM Journal on Mathematical Analysis

JF - SIAM Journal on Mathematical Analysis

IS - 1

ER -