Mathematics > Optimization and Control
[Submitted on 11 Feb 2020 (this version), latest version 21 Jan 2021 (v4)]
Title:Universal Average-Case Optimality of Polyak Momentum
View PDFAbstract:We consider the average-case runtime analysis of algorithms for minimizing quadratic objectives. In this setting, and contrary to the more classical worst-case analysis, non-asymptotic convergence rates and optimal algorithms depend on the full spectrum of the Hessian through its expected spectral distribution. Under mild assumptions, we show that these optimal methods converge asymptotically towards Polyak momentum independently of the expected spectral density. This makes Polyak momentum universally (i.e., independent of the spectral distribution) asymptotically average-case optimal.
Submission history
From: Damien Scieur [view email][v1] Tue, 11 Feb 2020 20:29:28 UTC (1,569 KB)
[v2] Wed, 1 Jul 2020 16:41:11 UTC (1,993 KB)
[v3] Wed, 19 Aug 2020 18:21:49 UTC (1,994 KB)
[v4] Thu, 21 Jan 2021 21:01:23 UTC (1,994 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.