GDDA: Semantic OOD Detection on Graphs under Covariate Shift via Score-Based Diffusion Models

TL;DR

This paper introduces GDDA, a novel framework for detecting out-of-distribution graphs under combined semantic and covariate shifts using score-based diffusion models, improving accuracy on benchmark datasets.

Contribution

The paper proposes a two-phase framework that disentangles graph representations and employs a diffusion model to generate out-of-distribution samples, addressing a previously under-explored problem.

Findings

Outperforms state-of-the-art OOD detection methods on benchmark datasets.

Effectively disentangles semantic and style factors in graph representations.

Generates realistic OOD samples to enhance detection accuracy.

Abstract

Out-of-distribution (OOD) detection poses a significant challenge for Graph Neural Networks (GNNs), particularly in open-world scenarios with varying distribution shifts. Most existing OOD detection methods on graphs primarily focus on identifying instances in test data domains caused by either semantic shifts (changes in data classes) or covariate shifts (changes in data features), while leaving the simultaneous occurrence of both distribution shifts under-explored. In this work, we address both types of shifts simultaneously and introduce a novel challenge for OOD detection on graphs: graph-level semantic OOD detection under covariate shift. In this scenario, variations between the training and test domains result from the concurrent presence of both covariate and semantic shifts, where only graphs associated with unknown classes are identified as OOD samples (OODs). To tackle this challenge, we propose a novel two-phase framework called Graph Disentangled Diffusion Augmentation (GDDA). The first phase focuses on disentangling graph representations into domain-invariant semantic factors and domain-specific style factors. In the second phase, we introduce a novel distribution-shift-controlled score-based generative diffusion model that generates latent factors outside the training semantic and style spaces. Additionally, auxiliary pseudo-in-distribution (InD) and pseudo-OOD graph representations are employed to enhance the effectiveness of the energy-based semantic OOD detector. Extensive empirical studies on three benchmark datasets demonstrate that our approach outperforms state-of-the-art baselines.