ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2205.15136
28
12

Optimal Gradient Sliding and its Application to Distributed Optimization Under Similarity

30 May 2022
D. Kovalev
Aleksandr Beznosikov
Ekaterina Borodich
Alexander Gasnikov
G. Scutari
ArXivPDFHTML
Abstract

We study structured convex optimization problems, with additive objective r:=p+qr:=p + qr:=p+q, where rrr is (μ\muμ-strongly) convex, qqq is LqL_qLq​-smooth and convex, and ppp is LpL_pLp​-smooth, possibly nonconvex. For such a class of problems, we proposed an inexact accelerated gradient sliding method that can skip the gradient computation for one of these components while still achieving optimal complexity of gradient calls of ppp and qqq, that is, O(Lp/μ)\mathcal{O}(\sqrt{L_p/\mu})O(Lp​/μ​) and O(Lq/μ)\mathcal{O}(\sqrt{L_q/\mu})O(Lq​/μ​), respectively. This result is much sharper than the classic black-box complexity O((Lp+Lq)/μ)\mathcal{O}(\sqrt{(L_p+L_q)/\mu})O((Lp​+Lq​)/μ​), especially when the difference between LqL_qLq​ and LqL_qLq​ is large. We then apply the proposed method to solve distributed optimization problems over master-worker architectures, under agents' function similarity, due to statistical data similarity or otherwise. The distributed algorithm achieves for the first time lower complexity bounds on {\it both} communication and local gradient calls, with the former having being a long-standing open problem. Finally the method is extended to distributed saddle-problems (under function similarity) by means of solving a class of variational inequalities, achieving lower communication and computation complexity bounds.

View on arXiv
Comments on this paper