FTF-ER: Feature-Topology Fusion-Based Experience Replay Method for Continual Graph Learning

TL;DR

This paper introduces FTF-ER, a novel experience replay method for continual graph learning that fuses feature and global topological information, improving efficiency and mitigating catastrophic forgetting.

Contribution

The paper proposes a feature-topology fusion approach with a new Hodge Potential Score for global topological importance, reducing memory and training costs in continual graph learning.

Findings

FTF-ER outperforms state-of-the-art methods with 3.6% higher AA and 7.1% higher AF on OGB-Arxiv.

HPS provides more accurate global topological importance than neighbor sampling.

The method reduces buffer storage and training time by excluding neighbor sampling.

Abstract

Continual graph learning (CGL) is an important and challenging task that aims to extend static GNNs to dynamic task flow scenarios. As one of the mainstream CGL methods, the experience replay (ER) method receives widespread attention due to its superior performance. However, existing ER methods focus on identifying samples by feature significance or topological relevance, which limits their utilization of comprehensive graph data. In addition, the topology-based ER methods only consider local topological information and add neighboring nodes to the buffer, which ignores the global topological information and increases memory overhead. To bridge these gaps, we propose a novel method called Feature-Topology Fusion-based Experience Replay (FTF-ER) to effectively mitigate the catastrophic forgetting issue with enhanced efficiency. Specifically, from an overall perspective to maximize the utilization of the entire graph data, we propose a highly complementary approach including both feature and global topological information, which can significantly improve the effectiveness of the sampled nodes. Moreover, to further utilize global topological information, we propose Hodge Potential Score (HPS) as a novel module to calculate the topological importance of nodes. HPS derives a global node ranking via Hodge decomposition on graphs, providing more accurate global topological information compared to neighbor sampling. By excluding neighbor sampling, HPS significantly reduces buffer storage costs for acquiring topological information and simultaneously decreases training time. Compared with state-of-the-art methods, FTF-ER achieves a significant improvement of 3.6% in AA and 7.1% in AF on the OGB-Arxiv dataset, demonstrating its superior performance in the class-incremental learning setting.