Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:1804.00325

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Machine Learning

arXiv:1804.00325 (cs)
[Submitted on 1 Apr 2018 (v1), last revised 1 May 2019 (this version, v3)]

Title:Aggregated Momentum: Stability Through Passive Damping

Authors:James Lucas, Shengyang Sun, Richard Zemel, Roger Grosse
View a PDF of the paper titled Aggregated Momentum: Stability Through Passive Damping, by James Lucas and 3 other authors
View PDF
Abstract:Momentum is a simple and widely used trick which allows gradient-based optimizers to pick up speed along low curvature directions. Its performance depends crucially on a damping coefficient $\beta$. Large $\beta$ values can potentially deliver much larger speedups, but are prone to oscillations and instability; hence one typically resorts to small values such as 0.5 or 0.9. We propose Aggregated Momentum (AggMo), a variant of momentum which combines multiple velocity vectors with different $\beta$ parameters. AggMo is trivial to implement, but significantly dampens oscillations, enabling it to remain stable even for aggressive $\beta$ values such as 0.999. We reinterpret Nesterov's accelerated gradient descent as a special case of AggMo and analyze rates of convergence for quadratic objectives. Empirically, we find that AggMo is a suitable drop-in replacement for other momentum methods, and frequently delivers faster convergence.
Comments: 11 primary pages, 11 supplementary pages, 12 figures total
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC); Machine Learning (stat.ML)
Cite as: arXiv:1804.00325 [cs.LG]
  (or arXiv:1804.00325v3 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.1804.00325
arXiv-issued DOI via DataCite
Journal reference: International Conference on Learning Representations, 2019

Submission history

From: James Lucas [view email]
[v1] Sun, 1 Apr 2018 17:53:03 UTC (3,415 KB)
[v2] Mon, 23 Jul 2018 16:49:44 UTC (3,324 KB)
[v3] Wed, 1 May 2019 14:09:52 UTC (3,244 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Aggregated Momentum: Stability Through Passive Damping, by James Lucas and 3 other authors
  • View PDF
  • TeX Source
  • Other Formats
view license
Current browse context:
cs.LG
< prev   |   next >
new | recent | 2018-04
Change to browse by:
cs
cs.AI
math
math.OC
stat
stat.ML

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar

DBLP - CS Bibliography

listing | bibtex
James Lucas
Richard S. Zemel
Richard Zemel
Roger Grosse
Roger B. Grosse
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
IArxiv Recommender (What is IArxiv?)
  • Author
  • Venue
  • Institution
  • Topic

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.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack