Passer à la navigation principale Passer à la recherche Passer au contenu principal

Unbalanced Optimal Transport through Non-negative Penalized Linear Regression

  • Laetitia Chapel
  • , Rémi Flamary
  • , Haoran Wu
  • , Cédric Févotte
  • , Gilles Gasso
  • Université Bretagne Sud
  • IRISA
  • Université Paul Sabatier
  • LITIS - Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collectionContribution à une conférenceRevue par des pairs

Résumé

This paper addresses the problem of Unbalanced Optimal Transport (UOT) in which the marginal conditions are relaxed (using weighted penalties in lieu of equality) and no additional regularization is enforced on the OT plan. In this context, we show that the corresponding optimization problem can be reformulated as a non-negative penalized linear regression problem. This reformulation allows us to propose novel algorithms inspired from inverse problems and nonnegative matrix factorization. In particular, we consider majorization-minimization which leads in our setting to efficient multiplicative updates for a variety of penalties. Furthermore, we derive for the first time an efficient algorithm to compute the regularization path of UOT with quadratic penalties. The proposed algorithm provides a continuity of piece-wise linear OT plans converging to the solution of balanced OT (corresponding to infinite penalty weights). We perform several numerical experiments on simulated and real data illustrating the new algorithms, and provide a detailed discussion about more sophisticated optimization tools that can further be used to solve OT problems thanks to our reformulation.

langue originaleAnglais
titreAdvances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021
rédacteurs en chefMarc'Aurelio Ranzato, Alina Beygelzimer, Yann Dauphin, Percy S. Liang, Jenn Wortman Vaughan
EditeurNeural information processing systems foundation
Pages23270-23282
Nombre de pages13
ISBN (Electronique)9781713845393
étatPublié - 1 janv. 2021
Evénement35th Conference on Neural Information Processing Systems, NeurIPS 2021 - Virtual, Online
Durée: 6 déc. 202114 déc. 2021

Série de publications

NomAdvances in Neural Information Processing Systems
Volume28
ISSN (imprimé)1049-5258

Une conférence

Une conférence35th Conference on Neural Information Processing Systems, NeurIPS 2021
La villeVirtual, Online
période6/12/2114/12/21

Empreinte digitale

Examiner les sujets de recherche de « Unbalanced Optimal Transport through Non-negative Penalized Linear Regression ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation