提出CNA模块缓解图神经网络过平滑问题，提升深层模型性能。
Graph Neural Networks Need Cluster-Normalize-Activate Modules
- 每层通过聚类-归一化-激活形成超节点，避免特征趋同。
- 在Cora和CiteSeer上准确率分别达94.18%和95.75%，优于现有方法。
- 参数量更少，适合资源受限场景下的深层图模型应用。
图神经网络（GNN）是处理图结构数据的非欧几里得深度学习模型。尽管应用广泛，但过平滑问题导致节点特征趋于同一固定点，严重限制了深层架构的性能。为此，我们提出一种即插即用的三步模块：聚类-归一化-激活（CNA）。通过在每层引入CNA模块，GNN可自动发现并形成超节点，对各超节点独立进行归一化与激活。实验表明，在节点分类与属性预测任务中，CNA显著提升性能。特别地，在Cora和CiteSeer数据集上准确率分别达到94.18%和95.75%。此外，该模块在回归任务中也表现优异，均方误差低于所有基线方法。同时，采用CNA的GNN所需可学习参数远少于其他先进架构。
原文摘要 · Abstract (English)
Graph Neural Networks (GNNs) are non-Euclidean deep learning models for graph-structured data. Despite their successful and diverse applications, oversmoothing prohibits deep architectures due to node features converging to a single fixed point. This severely limits their potential to solve complex tasks. To counteract this tendency, we propose a plug-and-play module consisting of three steps: Cluster-Normalize-Activate (CNA). By applying CNA modules, GNNs search and form super nodes in each layer, which are normalized and activated individually. We demonstrate in node classification and property prediction tasks that CNA significantly improves the accuracy over the state-of-the-art. Particularly, CNA reaches 94.18% and 95.75% accuracy on Cora and CiteSeer, respectively. It further benefits GNNs in regression tasks as well, reducing the mean squared error compared to all baselines. At the same time, GNNs with CNA require substantially fewer learnable parameters than competing architectures.
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。