PuSH - Publikationsserver des Helmholtz Zentrums München

Export: TXT Text RIS Endnote (RIS) BIB BibTeX
Scetbon, M.* ; Klein, M.* ; Palla, G. ; Cuturi, M.C.*

Unbalanced Low-rank Optimal Transport Solvers.

In: (37th Conference on Neural Information Processing Systems (NeurIPS), 10-16 December 2023, New Orleans, LA). 10010 North Torrey Pines Rd, La Jolla, California 92037 Usa: Neural Information Processing Systems (nips), 2023. 14
Two salient limitations have long hindered the relevance of optimal transport methods to machine learning. First, the O(n(3)) computational cost of standard sample-based solvers (when used on batches of n samples) is prohibitive. Second, the mass conservation constraint makes OT solvers too rigid in practice: because they must match all points from both measures, their output can be heavily influenced by outliers. A flurry of recent works has addressed these computational and modeling limitations, but has resulted in two separate strains of methods: While the computational outlook was much improved by entropic regularization, more recent O(n) linear-time low-rank solvers hold the promise to scale up OT further. In terms of modeling flexibility, the rigidity of mass conservation has been eased for entropic regularized OT, thanks to unbalanced variants of OT that can penalize couplings whose marginals deviate from those specified by the source and target distributions. The goal of this paper is to merge these two strains, low-rank and unbalanced, to achieve the promise of solvers that are both scalable and versatile. We propose custom algorithms to implement these extensions for the linear OT problem and its fused-Gromov-Wasserstein generalization, and demonstrate their practical relevance to challenging spatial transcriptomics matching problems. These algorithms are implemented in the ott-jax toolbox [Cuturi et al., 2022].
Tags
Anmerkungen
Besondere Publikation
Auf Hompepage verbergern

Zusatzinfos bearbeiten
Eigene Tags bearbeiten
Privat
Eigene Anmerkung bearbeiten
Privat
Auf Publikationslisten für
Homepage nicht anzeigen
Als besondere Publikation
markieren
Publikationstyp Artikel: Konferenzbeitrag
Schlagwörter Convergence; Algorithm
Sprache englisch
Veröffentlichungsjahr 2023
HGF-Berichtsjahr 2023
ISSN (print) / ISBN 1049-5258
Konferenztitel 37th Conference on Neural Information Processing Systems (NeurIPS)
Konferzenzdatum 10-16 December 2023
Konferenzort New Orleans, LA
Quellenangaben Band: , Heft: , Seiten: 14 Artikelnummer: , Supplement: ,
Verlag Neural Information Processing Systems (nips)
Verlagsort 10010 North Torrey Pines Rd, La Jolla, California 92037 Usa
Institut(e) Institute of Computational Biology (ICB)
POF Topic(s) 30205 - Bioengineering and Digital Health
Forschungsfeld(er) Enabling and Novel Technologies
PSP-Element(e) G-503800-001
WOS ID 001230083401036
Scopus ID 85205445005
Erfassungsdatum 2024-07-30
[➜Korrektur melden]