Metraipolis › archive › dc_74986d36a49e

Projected Riemannian Gradient Descent for the Bures-Wasserstein Barycenter: Dimension-Independent Linear Convergence at Unit Step Size — arXiv:2609.03762

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.03762
License (as archived)
CC0-1.0
Archived
2026-10-17T02:08:09.075Z
SHA-256 of the archived text
1d388cac22b97386d621ba4f1f032eef10c6742f61cfee8da05d1dfa065f569d
Stable URL of this page
https://metraipolis.com/documento/dc_74986d36a49e

How to cite

Projected Riemannian Gradient Descent for the Bures-Wasserstein Barycenter: Dimension-Independent Linear Convergence at Unit Step Size — arXiv:2609.03762 — arXiv (en), https://arxiv.org/abs/2609.03762 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_74986d36a49e (Metraipolis archive dc_74986d36a49e, sha256 1d388cac22b97386d621ba4f1f032eef10c6742f61cfee8da05d1dfa065f569d)
BibTeX
@misc{dc74986d36a49e,
  title = {Projected Riemannian Gradient Descent for the Bures-Wasserstein Barycenter: Dimension-Independent Linear Convergence at Unit Step Size — arXiv:2609.03762},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03762},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_74986d36a49e; sha256 1d388cac22b97386d621ba4f1f032eef10c6742f61cfee8da05d1dfa065f569d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03762. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The computation of the Bures-Wasserstein (BW) barycenter of an ensemble of positive definite matrices arises throughout machine learning, optimal transport, and quantum information. Riemannian gradient descent (RGD) at unit step size -- the fixed-point iteration used in practice -- converges rapidly, yet existing analyses present a dichotomy: unit-step guarantees carry worst-case exponential dependence on the dimension, while dimension-independent guarantees require small step sizes that forfeit the empirical speed. We resolve this dichotomy, not by improving the guarantees for unit-step RGD, but by proposing a Projected RGD algorithm that achieves dimension-independent linear convergence at unit step size. The achieved rate, $(1 - \kappa^{-3/2})$, where $\kappa$ is the condition number of the ensemble, also polynomially improves on the best small-step guarantee ($\kappa^{3/2}$ versus $\kappa^{5/2}$ iteration complexity). The crux is a novel Projection Lemma: clipping the eigenvalues of a positive matrix to an interval $[\alpha, \beta]$ is the closed-form, non-expansive (1-Lipschitz) BW-metric projection onto the set $\{S : \alpha I \leq S \leq \beta I\}$ -- a statement which, unlike its known one-sided counterpart, does not follow from convexity. The projection is moreover free: it reuses an eigendecomposition the next iteration must perform in any case, so the projected and unprojected iterations cost the same per step. The same analysis covers the invariant matrix projection problem of Brahmachari et al. (2025), whose fixed-point algorithm we identify as unit-step RGD on a totally geodesic submanifold, thereby extending the dimension-independent guarantee to that setting verbatim.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Optimization and Control (math.OC); Quantum Physics (quant-ph).
Pagina del lavoro: https://arxiv.org/abs/2609.03762 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T01:47:25.564Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).