A new flow-based algorithm solves the data-driven Monge optimal transport barycenter problem by recasting the independence constraint as a singular-value penalty, yielding linear per-iteration cost and a closed-form map inverse.
Barycenter in the Wasserstein space.SIAM J
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
math.OC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
The Monge optimal transport barycenter problem
A new flow-based algorithm solves the data-driven Monge optimal transport barycenter problem by recasting the independence constraint as a singular-value penalty, yielding linear per-iteration cost and a closed-form map inverse.