cs.LGJul 17, 2026

Hierarchical Domain Generalization

Authors: Chenxiao YangZhiyuan LiShai Ben-DavidNathan Srebro

Organizations: Toyota Technological Institute at Chicago · University of Waterloo

Abstract

We study hierarchical domain generalization as a problem of extrapolation from finite observed regions to an entire instance space, replacing i.i.d. sampling with arbitrary domain hierarchies. We show that the central obstruction is not only the complexity of the hypothesis class, but the train/test domain partition through which evidence is revealed. In particular, no matter how small the class or how large the training size, some partition makes generalization fail for some target. These results suggest that modern generalization theory must treat domain structure as a first-class object.

Explore similar work

CardsList