Salvato in:
Dettagli Bibliografici
Autori principali: Shu, Yunlu, Jiang, Jiaxin, Shi, Lei, Wang, Tianyu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2504.06814
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918161361666048
author Shu, Yunlu
Jiang, Jiaxin
Shi, Lei
Wang, Tianyu
author_facet Shu, Yunlu
Jiang, Jiaxin
Shi, Lei
Wang, Tianyu
contents In a seminal work of Zhang and Sra, gradient descent methods for geodesically convex optimization were comprehensively studied. In particular, Zhang and Sra derived a comparison inequality that relates the iterative points in the optimization process. Since their seminal work, numerous follow-ups have studied different downstream usages of their comparison lemma. In this work, we introduce the concept of quasilinearization to optimization, presenting a novel framework for analyzing geodesically convex optimization. By leveraging this technique, we establish state-of-the-art convergence rates -- for both deterministic and stochastic settings -- under weaker assumptions than previously required. The technique of quasilinearization may prove valuable for other non-Euclidean optimization problems.
format Preprint
id arxiv_https___arxiv_org_abs_2504_06814
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Revisit First-order Methods for Geodesically Convex Optimization
Shu, Yunlu
Jiang, Jiaxin
Shi, Lei
Wang, Tianyu
Optimization and Control
In a seminal work of Zhang and Sra, gradient descent methods for geodesically convex optimization were comprehensively studied. In particular, Zhang and Sra derived a comparison inequality that relates the iterative points in the optimization process. Since their seminal work, numerous follow-ups have studied different downstream usages of their comparison lemma. In this work, we introduce the concept of quasilinearization to optimization, presenting a novel framework for analyzing geodesically convex optimization. By leveraging this technique, we establish state-of-the-art convergence rates -- for both deterministic and stochastic settings -- under weaker assumptions than previously required. The technique of quasilinearization may prove valuable for other non-Euclidean optimization problems.
title Revisit First-order Methods for Geodesically Convex Optimization
topic Optimization and Control
url https://arxiv.org/abs/2504.06814