Learning Invariant Graph Representations Through Redundant Information
By: Barproda Halder, Pasan Dissanayake, Sanghamitra Dutta
Potential Business Impact:
Helps computers learn from new data without mistakes.
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 $Y$ shared between spurious subgraphs $G_s$ and invariant subgraphs $G_c$ 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.
Similar Papers
Redundancy-Aware Test-Time Graph Out-of-Distribution Detection
Machine Learning (CS)
Helps computers spot fake data in graphs.
PISA: Prioritized Invariant Subgraph Aggregation
Machine Learning (CS)
Finds better patterns in complex data for more accurate predictions.
Pseudo-label Induced Subspace Representation Learning for Robust Out-of-Distribution Detection
Machine Learning (CS)
Helps AI spot fake or new information.