cs.LGNov 6, 2026

Learning Across Heterogeneous Feature Spaces with Group Robustness

Authors: Xenia Konti, Arjun Tschand, Chuan Hong, Nicoleta J. Economou-Zavlanos, Michael M. Zavlanos

Organizations: Duke University

Abstract

In clinical applications, patients undergo different combinations of tests, and machine learning models must therefore ensure accurate predictions across heterogeneous feature spaces. Current approaches upweight the patients with the highest loss, without accounting for the limited predictive information in their features preventing their loss from going any lower. Moreover, when patients' feature spaces are fully disjoint, current methods reduce to training independent models, preventing knowledge transfer across them. We propose a framework for group robust learning across heterogeneous feature spaces, where groups are patients sharing a feature space. Since a high loss may simply reflect uninformative features, we instead optimize for worst-group excess loss, the gap between a group's loss and an estimate of its achievable optimum. To allow collaboration between groups, we use group-specific encoders mapping into a shared latent space in which learnable Gaussian anchors cluster patients by outcome. On tabular and imaging medical datasets, including fully disjoint feature settings, our framework outperforms state-of-the-art methods when groups differ in their optimal losses or cannot train a good model on their own, reducing worst-group loss by up to 79% and excess loss by up to 95% with up to 18×\times fewer parameters.

Figures & tables

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Rank-Constrained Adaptation for Reliable Real-World Performance

    Feb 6, 2026Abinitha Gourabathina, Hyewon Jeong, Teya Bergamaschi +2Subgroup RobustnessDistribution Shift Robustness

  2. ProME: Prototype-Margin Environments with Repair-Aware Selection for Group-Robust Learning

    Aug 13, 2026Qianqian Wang, Yunshan Li, Dawei Huang +2Distribution Shift RobustnessSpurious Correlation Robustness

  3. Automatic Discovery of Disease Subgroups by Contrasting with Healthy Controls

    May 20, 2026Robin Louiset, Edouard Duchesnay, Benoit Dufumier +2Contrastive LearningHealthcare