AC2L-GAD: Active Counterfactual Contrastive Learning for Graph Anomaly Detection

TL;DR

AC2L-GAD introduces an active counterfactual contrastive learning framework for graph anomaly detection, effectively addressing label scarcity and class imbalance by generating informative positive and negative samples through principled counterfactual reasoning.

Contribution

It presents a novel active counterfactual contrastive learning method that improves anomaly detection in graphs by reducing computational costs and enhancing contrast quality.

Findings

Achieves competitive or superior performance on nine benchmark datasets.

Reduces computational overhead by approximately 65% compared to full counterfactual generation.

Shows notable gains in detecting complex anomalies with attribute-structure interactions.

Abstract

Graph anomaly detection aims to identify abnormal patterns in networks, but faces significant challenges from label scarcity and extreme class imbalance. While graph contrastive learning offers a promising unsupervised solution, existing methods suffer from two critical limitations: random augmentations break semantic consistency in positive pairs, while naive negative sampling produces trivial, uninformative contrasts. We propose AC2L-GAD, an Active Counterfactual Contrastive Learning framework that addresses both limitations through principled counterfactual reasoning. By combining information-theoretic active selection with counterfactual generation, our approach identifies structurally complex nodes and generates anomaly-preserving positive augmentations alongside normal negative counterparts that provide hard contrasts, while restricting expensive counterfactual generation to a strategically selected subset. This design reduces computational overhead by approximately 65% compared to full-graph counterfactual generation while maintaining detection quality. Experiments on nine benchmark datasets, including real-world financial transaction graphs from GADBench, show that AC2L-GAD achieves competitive or superior performance compared to state-of-the-art baselines, with notable gains in datasets where anomalies exhibit complex attribute-structure interactions.