DiffGAD: A Diffusion-based Unsupervised Graph Anomaly Detector

2024-10-09

arXiv.org

AI Summary

Traditional unsupervised graph anomaly detection methods rely on reconstruction error but struggle to capture discriminative structural features, limiting their anomaly identification capability. To address this, we propose DiffGAD—the first framework to introduce denoising diffusion probabilistic models (DDPMs) into graph anomaly detection. DiffGAD jointly models node-level local structures and global topological semantics via multi-step progressive denoising in a learned latent space. It further introduces a novel cross-scale content preservation mechanism that enhances anomaly-sensitive feature representation with low computational and memory overhead. Crucially, DiffGAD is fully unsupervised: it requires neither labels nor anomaly priors, training exclusively on normal graph data. Evaluated on six large-scale real-world graph datasets, DiffGAD consistently outperforms existing state-of-the-art methods across standard metrics—including AUC and Average Precision—with average improvements of 3.2–7.8 percentage points.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Computer Vision: Diffusion Models for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Graph Anomaly Detection (GAD) is crucial for identifying abnormal entities within networks, garnering significant attention across various fields. Traditional unsupervised methods, which decode encoded latent representations of unlabeled data with a reconstruction focus, often fail to capture critical discriminative content, leading to suboptimal anomaly detection. To address these challenges, we present a Diffusion-based Graph Anomaly Detector (DiffGAD). At the heart of DiffGAD is a novel latent space learning paradigm, meticulously designed to enhance its proficiency by guiding it with discriminative content. This innovative approach leverages diffusion sampling to infuse the latent space with discriminative content and introduces a content-preservation mechanism that retains valuable information across different scales, significantly improving its adeptness at identifying anomalies with limited time and space complexity. Our comprehensive evaluation of DiffGAD, conducted on six real-world and large-scale datasets with various metrics, demonstrated its exceptional performance.

Problem

• Enhances unsupervised graph anomaly detection.
• Leverages diffusion for discriminative content.
• Reduces time and space complexity.

Innovation

• Diffusion-based latent space learning
• Discriminative content-guided enhancement
• Content-preservation across different scales