Metraipolis › archive › dc_4eb131bd012d
Adaptive Bregman Alternating Projections for Feasible Gromov-Wasserstein Learning — arXiv:2610.04264
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04264
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:30:29.980Z
- SHA-256 of the archived text
dc93a6ae662f9c3eb0a029ca67b1ca7e559b2cb0ec34997c0f5f14720dc243ae- Stable URL of this page
https://metraipolis.com/documento/dc_4eb131bd012d
How to cite
Adaptive Bregman Alternating Projections for Feasible Gromov-Wasserstein Learning — arXiv:2610.04264 — arXiv, https://arxiv.org/abs/2610.04264 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4eb131bd012d (Metraipolis archive dc_4eb131bd012d, sha256 dc93a6ae662f9c3eb0a029ca67b1ca7e559b2cb0ec34997c0f5f14720dc243ae)
BibTeX
@misc{dc4eb131bd012d,
title = {Adaptive Bregman Alternating Projections for Feasible Gromov-Wasserstein Learning — arXiv:2610.04264},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04264},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4eb131bd012d; sha256 dc93a6ae662f9c3eb0a029ca67b1ca7e559b2cb0ec34997c0f5f14720dc243ae},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04264. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
The Gromov-Wasserstein (GW) problem compares structured distributions without requiring a shared feature space or known correspondences, but its nonconvex objective and coupled marginal constraints make computation challenging. Bregman alternating projected gradient (BAPG) uses inexpensive alternating row and column updates, yet its fixed-penalty relaxation leaves a persistent feasibility gap. We propose Adaptive KL-BAPG (A-KL-BAPG), which combines a finite fixed-penalty burn-in with a guarded increasing-penalty phase. At each tail iteration, the method reuses BAPG's alternating updates and backtracks a delayed-power step until a Sinkhorn-inspired projective-diameter safeguard is satisfied. We prove finite termination of the backtracking at each iteration and show that the feasibility gap vanishes asymptotically. We further establish a best-iterate $O(1/\log N)$ bound for the weighted squared corrected residual and, under a support regularity condition, the existence of a stationary accumulation point for the original GW problem. This distinguishes A-KL-BAPG from fixed-penalty BAPG, whose stationarity guarantees are given for the relaxed problem. Experiments show that A-KL-BAPG achieves a favorable balance of accuracy, objective value, feasibility, and stationarity relative to BAPG variants, projection-based methods, and task-specific baselines. For synthetic and real graph alignment problems, it closely matches the accuracy and objective value of fixed-penalty KL-BAPG while reducing the marginal feasibility gap by 62-99% and the projected stationarity residual by 28-98%. Heterogeneous domain adaptation experiments show a similar pattern: A-KL-BAPG maintains comparable target accuracy and objective values while achieving better feasibility and stationarity than fixed-penalty KL-BAPG. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Computational Geometry (cs.CG); Optimization and Control (math.OC). Pagina del lavoro: https://arxiv.org/abs/2610.04264 — 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 2026-10-06T05:27:05.669Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).