Corpus ID: 211082751

Revisiting Fixed Support Wasserstein Barycenter: Computational Hardness and Efficient Algorithms

@article{Lin2020RevisitingFS,
  title={Revisiting Fixed Support Wasserstein Barycenter: Computational Hardness and Efficient Algorithms},
  author={Tianyi Lin and Nhat Ho and X. Chen and Marco Cuturi and Michael I. Jordan},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.04783}
}
We study the fixed-support Wasserstein barycenter problem (FS-WBP), which consists in computing the Wasserstein barycenter of m discrete probability measures supported on a finite metric space of size n. We show first that the constraint matrix arising from the linear programming (LP) representation of the FS-WBP is totally unimodular when m ≥ 3 and n = 2, but not totally unimodular when m ≥ 3 and n ≥ 3. This result answers an open problem, since it shows that the FS-WBP is not a minimum-cost… Expand
4 Citations

References

SHOWING 1-10 OF 69 REFERENCES
A Fast Globally Linearly Convergent Algorithm for the Computation of Wasserstein Barycenters
  • 11
  • Highly Influential
  • PDF
On the Complexity of Approximating Wasserstein Barycenters
  • 26
  • Highly Influential
  • PDF
Discrete Wasserstein barycenters: optimal transport for discrete data
  • 59
  • Highly Influential
  • PDF
Fast Computation of Wasserstein Barycenters
  • 413
  • PDF
Interior-Point Methods Strike Back: Solving the Wasserstein Barycenter Problem
  • 10
  • Highly Influential
  • PDF
Improved Linear Programs for Discrete Barycenters
  • 10
  • PDF
A Direct Õ(1/ε) Iteration Parallel Algorithm for Optimal Transport
  • 16
  • Highly Influential
  • PDF
Decentralize and Randomize: Faster Algorithm for Wasserstein Barycenters
  • 56
  • Highly Influential
  • PDF
On the Computation of Wasserstein Barycenters
  • 8
Fast Discrete Distribution Clustering Using Wasserstein Barycenter With Sparse Support
  • 68
  • Highly Influential
  • PDF
...
1
2
3
4
5
...