Evolving Graph Learning for Out-of-Distribution Generalization in Non-stationary Environments.

2025-11-04

IEEE Transactions on Pattern Analysis and Machine Intelligence

AI Summary

To address distribution shift and degraded out-of-distribution (OOD) generalization in dynamic graphs under non-stationary environments caused by environmental evolution, this paper proposes EvoGOOD—a novel framework for OOD generalization on dynamic graphs. EvoGOOD is the first to model non-stationarity from an environmental evolution perspective: it introduces an environmental sequence variational autoencoder to characterize the temporal evolution of environmental distributions, and incorporates environment-aware invariant learning. Furthermore, it proposes a node-level causal intervention method based on environmental inference and instance mixing, enabling fine-grained identification and disentanglement of spatiotemporally invariant features. Extensive experiments on multiple real-world and synthetic dynamic graph datasets demonstrate that EvoGOOD significantly improves model robustness and predictive performance under OOD scenarios. The framework establishes a new paradigm for OOD generalization in dynamic graph learning.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Graph neural networks have shown remarkable success in exploiting the spatial and temporal patterns on dynamic graphs. However, existing GNNs exhibit poor generalization ability under distribution shifts, which is inevitable in dynamic scenarios. As dynamic graph generation progresses amid evolving latent non-stationary environments, it is imperative to explore their effects on out-of-distribution (OOD) generalization. This paper proposes a novel Evolving Graph Learning framework for OOD generalization (EvoGOOD) by environment-aware invariant pattern recognition. Specifically, we first design an environment sequential variational auto-encoder to model environment evolution and infer underlying environment distribution. Then, we introduce a mechanism for environment-aware invariant pattern recognition, tailored to address environmental diversification through inferred distributions. Finally, we conduct fine-grained causal interventions on individual nodes using a mixture of instantiated environment samples. This approach helps to distinguish spatio-temporal invariant patterns for OOD prediction, especially in non-stationary environments. Experimental results demonstrate the superiority of EvoGOOD on both real-world and synthetic dynamic datasets under distribution shifts. To the best of our knowledge, it is the first attempt to study the dynamic graph OOD generalization problem from the environment evolution perspective.

Problem

• Addressing poor generalization of GNNs under distribution shifts in dynamic graphs
• Modeling environment evolution for out-of-distribution prediction in non-stationary scenarios
• Identifying spatio-temporal invariant patterns through environment-aware causal interventions

Innovation

• Environment-aware invariant pattern recognition for graphs
• Sequential variational auto-encoder models environment evolution
• Fine-grained causal interventions on nodes using environment samples