Robustness to distribution shifts is critical for deploying machine learning models in the real
world. Despite this necessity, there has been little work in defining the underlying …

Distribution shifts—where the training distribution differs from the test distribution—can
substantially degrade the accuracy of machine learning (ML) systems deployed in the wild …

Different distribution shifts require different algorithmic and operational interventions.
Methodological research must be grounded by the specific shifts they address. Although …

Generalization under distributional shift is an open challenge for machine learning. Invariant
Risk Minimization (IRM) is a promising framework to tackle this issue by extracting invariant …

We develop a methodology for assessing the robustness of models to subpopulation shift---
specifically, their ability to generalize to novel data subpopulations that were not observed …

Understanding the performance of machine learning models across diverse data
distributions is critically important for reliable applications. Motivated by this, there is a …

We often see undesirable tradeoffs in robust machine learning where out-of-distribution
(OOD) accuracy is at odds with in-distribution (ID) accuracy. A robust classifier obtained via …

A fundamental assumption of most machine learning algorithms is that the training and test
data are drawn from the same underlying distribution. However, this assumption is violated …

A central goal of machine learning is to learn robust representations that capture the
fundamental relationship between inputs and output labels. However, minimizing training …

We introduce four new real-world distribution shift datasets consisting of changes in image
style, image blurriness, geographic location, camera operation, and more. With our new …