Abstract:We consider learning from labeled data collected across multiple environments, where the data distribution may vary across these environments. This problem is commonly approached from a causal perspective, seeking invariant representations that retain causal factors while discarding spurious ones. However, this framework assumes that the environment has no direct effect on the target. In contrast, we consider settings in which this assumption fails, but still aim to learn representations that support robust prediction on average across previously unseen environments. To this end, we study representations learned by explicitly modeling variation across environments and then marginalizing that variation out. We analyze the resulting representations and characterize when they are preferable to those learned by causal invariant-representation methods. We propose a concrete method based on generalized random-intercept models, a class of predictors in which such marginalization is possible, and study their generalization properties. Empirically, we show that these models outperform invariant-learning methods across a range of challenging settings.
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG) |
| Cite as: | arXiv:2604.26128 [stat.ML] |
| (or arXiv:2604.26128v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2604.26128 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuli Slavutsky [view email]
[v1]
Tue, 28 Apr 2026 21:36:56 UTC (148 KB)