Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:1912.10095

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Machine Learning

arXiv:1912.10095 (cs)
[Submitted on 20 Dec 2019 (v1), last revised 23 Jul 2020 (this version, v2)]

Title:Landscape Connectivity and Dropout Stability of SGD Solutions for Over-parameterized Neural Networks

Authors:Alexander Shevchenko, Marco Mondelli
View a PDF of the paper titled Landscape Connectivity and Dropout Stability of SGD Solutions for Over-parameterized Neural Networks, by Alexander Shevchenko and Marco Mondelli
View PDF
Abstract:The optimization of multilayer neural networks typically leads to a solution with zero training error, yet the landscape can exhibit spurious local minima and the minima can be disconnected. In this paper, we shed light on this phenomenon: we show that the combination of stochastic gradient descent (SGD) and over-parameterization makes the landscape of multilayer neural networks approximately connected and thus more favorable to optimization. More specifically, we prove that SGD solutions are connected via a piecewise linear path, and the increase in loss along this path vanishes as the number of neurons grows large. This result is a consequence of the fact that the parameters found by SGD are increasingly dropout stable as the network becomes wider. We show that, if we remove part of the neurons (and suitably rescale the remaining ones), the change in loss is independent of the total number of neurons, and it depends only on how many neurons are left. Our results exhibit a mild dependence on the input dimension: they are dimension-free for two-layer networks and depend linearly on the dimension for multilayer networks. We validate our theoretical findings with numerical experiments for different architectures and classification tasks.
Comments: Proceedings of the 37th International Conference on Machine Learning (ICML)
Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
Cite as: arXiv:1912.10095 [cs.LG]
  (or arXiv:1912.10095v2 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.1912.10095
arXiv-issued DOI via DataCite

Submission history

From: Alexander Shevchenko [view email]
[v1] Fri, 20 Dec 2019 20:49:52 UTC (633 KB)
[v2] Thu, 23 Jul 2020 13:26:37 UTC (199 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Landscape Connectivity and Dropout Stability of SGD Solutions for Over-parameterized Neural Networks, by Alexander Shevchenko and Marco Mondelli
  • View PDF
  • TeX Source
  • Other Formats
view license
Current browse context:
cs.LG
< prev   |   next >
new | recent | 2019-12
Change to browse by:
cs
stat
stat.ML

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar

DBLP - CS Bibliography

listing | bibtex
Marco Mondelli
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