Condensation-Concatenation Framework for Dynamic Graph Continual Learning
This addresses the problem of continual learning on dynamic graphs for applications like social networks or recommendation systems, representing an incremental improvement by refining methods for topological changes.
The paper tackles catastrophic forgetting in graph neural networks on dynamic graphs by proposing the Condensation-Concatenation-based Continual Learning (CCC) framework, which condenses historical snapshots and concatenates embeddings to preserve label distribution and topology, achieving superior performance over state-of-the-art baselines on four real-world datasets.
Dynamic graphs are prevalent in real-world scenarios, where continuous structural changes induce catastrophic forgetting in graph neural networks (GNNs). While continual learning has been extended to dynamic graphs, existing methods overlook the effects of topological changes on existing nodes. To address it, we propose a novel framework for continual learning on dynamic graphs, named Condensation-Concatenation-based Continual Learning (CCC). Specifically, CCC first condenses historical graph snapshots into compact semantic representations while aiming to preserve the original label distribution and topological properties. Then it concatenates these historical embeddings with current graph representations selectively. Moreover, we refine the forgetting measure (FM) to better adapt to dynamic graph scenarios by quantifying the predictive performance degradation of existing nodes caused by structural updates. CCC demonstrates superior performance over state-of-the-art baselines across four real-world datasets in extensive experiments.