提出聚合缓冲区，提升图神经网络在删边增强下的鲁棒性
Aggregation Buffer: Revisiting DropEdge with a New Parameter Block
- 设计参数模块缓解删边增强的固有缺陷
- 在多个数据集上实现一致性能提升
- 统一解决度偏差与结构不均衡问题
我们重新审视了用于图神经网络的数据增强技术DropEdge，该方法通过随机删除边来暴露训练期间多样的图结构。尽管能有效减少对特定连接的过拟合，但其在监督学习任务中的性能提升有限。我们通过理论分析发现，这一局限源于许多图神经网络架构的根本缺陷。基于此，我们提出聚合缓冲区（Aggregation Buffer），一个专门设计的参数块，旨在通过解决DropEdge的局限性来增强图神经网络的鲁棒性。该方法兼容任意图神经网络模型，并在多个数据集上均表现出一致的性能提升。此外，该方法作为一种统一方案，有效缓解了度偏差和结构差异等经典问题。代码与数据集见https://github.com/dooho00/agg-buffer。
原文摘要 · Abstract (English)
We revisit DropEdge, a data augmentation technique for GNNs which randomly removes edges to expose diverse graph structures during training. While being a promising approach to effectively reduce overfitting on specific connections in the graph, we observe that its potential performance gain in supervised learning tasks is significantly limited. To understand why, we provide a theoretical analysis showing that the limited performance of DropEdge comes from the fundamental limitation that exists in many GNN architectures. Based on this analysis, we propose Aggregation Buffer, a parameter block specifically designed to improve the robustness of GNNs by addressing the limitation of DropEdge. Our method is compatible with any GNN model, and shows consistent performance improvements on multiple datasets. Moreover, our method effectively addresses well-known problems such as degree bias or structural disparity as a unifying solution. Code and datasets are available at https://github.com/dooho00/agg-buffer.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。