Corpus ID: 195820364

Invariant Risk Minimization

@article{Arjovsky2019InvariantRM,
  title={Invariant Risk Minimization},
  author={Mart{\'i}n Arjovsky and L. Bottou and Ishaan Gulrajani and David Lopez-Paz},
  journal={ArXiv},
  year={2019},
  volume={abs/1907.02893}
}
We introduce Invariant Risk Minimization (IRM), a learning paradigm to estimate invariant correlations across multiple training distributions. To achieve this goal, IRM learns a data representation such that the optimal classifier, on top of that data representation, matches for all training distributions. Through theory and experiments, we show how the invariances learned by IRM relate to the causal structures governing the data and enable out-of-distribution generalization. 
201 Citations
An Empirical Study of Invariant Risk Minimization
  • 9
  • Highly Influenced
  • PDF
The Risks of Invariant Risk Minimization
  • 15
  • Highly Influenced
  • PDF
Invariant Risk Minimization Games
  • 28
  • Highly Influenced
  • PDF
Learning Robust Models Using The Principle of Independent Causal Mechanisms
  • Highly Influenced
  • PDF
Risk Variance Penalization: From Distributional Robustness to Causality
  • 6
  • Highly Influenced
  • PDF
Invariant Adversarial Learning for Distributional Robustness
  • Highly Influenced
  • PDF
Out-of-Distribution Generalization via Risk Extrapolation (REx)
  • 56
  • Highly Influenced
  • PDF
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 58 REFERENCES
Principles of Risk Minimization for Learning Theory
  • 486
  • PDF
Invariant Models for Causal Transfer Learning
  • 97
  • PDF
On causal and anticausal learning
  • 237
  • Highly Influential
  • PDF
Robust Supervised Learning
  • 28
  • PDF
Stable Prediction across Unknown Environments
  • 35
  • PDF
Learning Causal Structures Using Regression Invariance
  • 24
  • PDF
Analysis of Representations for Domain Adaptation
  • 1,049
  • PDF
Deep Domain Generalization via Conditional Invariant Adversarial Networks
  • 125
  • PDF
Statistical learning theory
  • 25,658
Invariant Causal Prediction for Nonlinear Models
  • 73
  • PDF
...
1
2
3
4
5
...