Learning Invariant Graph Representations Through Redundant Information
Abstract
Learning invariant graph representations for out-of-distribution (OOD) generalization remains challenging because the learned representations often retain spurious components. To address this challenge, this work introduces a new tool from information theory called Partial Information Decomposition (PID) that goes beyond classical information-theoretic measures. We identify limitations in existing approaches for invariant representation learning that solely rely on classical information-theoretic measures, motivating the need to precisely focus on redundant information about the target shared between spurious subgraphs and invariant subgraphs obtained via PID. Next, we propose a new multi-level optimization framework that we call -- Redundancy-guided Invariant Graph learning (RIG) -- that maximizes redundant information while isolating spurious and causal subgraphs, enabling OOD generalization under diverse distribution shifts. Our approach relies on alternating between estimating a lower bound of redundant information (which itself requires an optimization) and maximizing it along with additional objectives. Experiments on both synthetic and real-world graph datasets demonstrate the generalization capabilities of our proposed RIG framework.
Cite
@article{arxiv.2512.06154,
title = {Learning Invariant Graph Representations Through Redundant Information},
author = {Barproda Halder and Pasan Dissanayake and Sanghamitra Dutta},
journal= {arXiv preprint arXiv:2512.06154},
year = {2025}
}