Invariant Link Selector for Spatial-Temporal Out-of-Distribution Problem

TL;DR

This paper introduces an Invariant Link Selector using the Information Bottleneck to identify invariant components in temporal graphs, enhancing model robustness against distribution shifts in real-world applications.

Contribution

It proposes a novel error-bounded invariant link selector that distinguishes invariant from variant components in temporal graphs for improved OOD generalization.

Findings

Outperforms state-of-the-art methods in temporal link prediction tasks.

Effectively identifies invariant components in temporal graphs.

Enhances model robustness in real-world applications like recommendation systems.

Abstract

In the era of foundation models, Out-of- Distribution (OOD) problems, i.e., the data discrepancy between the training environments and testing environments, hinder AI generalization. Further, relational data like graphs disobeying the Independent and Identically Distributed (IID) condition makes the problem more challenging, especially much harder when it is associated with time. Motivated by this, to realize the robust invariant learning over temporal graphs, we want to investigate what components in temporal graphs are most invariant and representative with respect to labels. With the Information Bottleneck (IB) method, we propose an error-bounded Invariant Link Selector that can distinguish invariant components and variant components during the training process to make the deep learning model generalizable for different testing scenarios. Besides deriving a series of rigorous generalizable optimization functions, we also equip the training with task-specific loss functions, e.g., temporal link prediction, to make pretrained models solve real-world application tasks like citation recommendation and merchandise recommendation, as demonstrated in our experiments with state-of-the-art (SOTA) methods. Our code is available at https://github.com/kthrn22/OOD-Linker.