Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > stat > arXiv:2101.01134

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Statistics > Machine Learning

arXiv:2101.01134 (stat)
[Submitted on 4 Jan 2021 (v1), last revised 26 Feb 2021 (this version, v2)]

Title:Does Invariant Risk Minimization Capture Invariance?

Authors:Pritish Kamath, Akilesh Tangella, Danica J. Sutherland, Nathan Srebro
View a PDF of the paper titled Does Invariant Risk Minimization Capture Invariance?, by Pritish Kamath and Akilesh Tangella and Danica J. Sutherland and Nathan Srebro
View PDF
Abstract:We show that the Invariant Risk Minimization (IRM) formulation of Arjovsky et al. (2019) can fail to capture "natural" invariances, at least when used in its practical "linear" form, and even on very simple problems which directly follow the motivating examples for IRM. This can lead to worse generalization on new environments, even when compared to unconstrained ERM. The issue stems from a significant gap between the linear variant (as in their concrete method IRMv1) and the full non-linear IRM formulation. Additionally, even when capturing the "right" invariances, we show that it is possible for IRM to learn a sub-optimal predictor, due to the loss function not being invariant across environments. The issues arise even when measuring invariance on the population distributions, but are exacerbated by the fact that IRM is extremely fragile to sampling.
Comments: Code is available in the arXiv ancillary files, linked from this page
Subjects: Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as: arXiv:2101.01134 [stat.ML]
  (or arXiv:2101.01134v2 [stat.ML] for this version)
  https://doi.org/10.48550/arXiv.2101.01134
arXiv-issued DOI via DataCite

Submission history

From: Danica J. Sutherland [view email]
[v1] Mon, 4 Jan 2021 18:02:45 UTC (2,644 KB)
[v2] Fri, 26 Feb 2021 23:21:48 UTC (4,785 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Does Invariant Risk Minimization Capture Invariance?, by Pritish Kamath and Akilesh Tangella and Danica J. Sutherland and Nathan Srebro
  • View PDF
  • TeX Source
  • Other Formats
view license
Ancillary-file links:

Ancillary files (details):

  • colored-mnist/main.py
  • colored-mnist/plot-results.ipynb
  • colored-mnist/run-experiments.sh
  • two-bit/pure-irm-fail-example-all.txt
  • two-bit/pure-irm-fail-example-no-theta.txt
  • two-bit/pure-irm-fail-example.py
  • two-bit/two-bit-irm.nb
  • two-bit/two-bit-irm.pdf
  • (3 additional files not shown)
Current browse context:
stat.ML
< prev   |   next >
new | recent | 2021-01
Change to browse by:
cs
cs.AI
cs.LG
stat

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
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?)
  • 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