Abstract
We study the barycenter of the Hellinger-Kantorovich metric over nonnegative measures on compact, convex subsets of \BbbR d. The article establishes existence, uniqueness (under suitable assumptions), and equivalence between a coupled-two-marginal and a multimarginal formulation. We analyze the HK barycenter between Dirac measures in detail, and find that it differs substantially from the Wasserstein barycenter by exhibiting a local ``clustering"" behavior, depending on the length scale of the input measures. In applications it makes sense to simultaneously consider all choices of this scale, leading to a 1-parameter family of barycenters. We demonstrate the usefulness of this family by analyzing point clouds sampled from a mixture of Gaussians and inferring the number and location of the underlying Gaussians.
Original language | English |
---|---|
Pages (from-to) | 62-110 |
Number of pages | 49 |
Journal | SIAM Journal on Mathematical Analysis |
Volume | 53 |
Issue number | 1 |
DOIs | |
State | Published - 2021 |
Keywords
- Barycenter
- Data analysis
- Multimarginal
- Optimal transport
- Unbalanced