DeepHGCN：迈向更深的双曲图卷积网络
DeepHGCN: Toward Deeper Hyperbolic Graph Convolutional Networks
AI总结：
本文提出首个深层多层双曲图卷积网络 DeepHGCN，通过快速准确的双曲特征变换、残差连接与正则化缓解计算开销和过平滑，并在链路预测与节点分类上显著优于欧几里得及浅层双曲基线。
摘要：
双曲图卷积网络（HGCN）在从层次化图中提取信息方面已展现出显著潜力。然而，由于双曲运算的计算开销以及随深度增加而出现的过平滑问题，现有 HGCN 局限于浅层架构。尽管已有一些处理方法被用于缓解 GCN 中的过平滑，但开发一种双曲解决方案面临独特挑战，因为必须谨慎设计运算以契合双曲特性。为应对这些挑战，我们提出 DeepHGCN，这是首个深度多层 HGCN 架构，可显著提升计算效率并大幅减轻过平滑。DeepHGCN 具有两项关键创新：（1）一种新型双曲特征变换层，可实现快速且准确的线性映射；（2）借助高效的双曲中点方法实现的双曲残差连接，以及针对权重和特征的正则化等技术。大量实验表明，与欧几里得 GCN 和浅层双曲 GCN 变体相比，DeepHGCN 在链路预测和节点分类任务上均取得显著提升。
英文摘要：
Hyperbolic graph convolutional networks (HGCNs) have demonstrated significant potential in extracting information from hierarchical graphs. However, existing HGCNs are limited to shallow architectures due to the computational expense of hyperbolic operations and the issue of over-smoothing as depth increases. Although treatments have been applied to alleviate over-smoothing in GCNs, developing a hyperbolic solution presents distinct challenges since operations must be carefully designed to fit the hyperbolic nature. Addressing these challenges, we propose DeepHGCN, the first deep multi-layer HGCN architecture with dramatically improved computational efficiency and substantially reduced over-smoothing. DeepHGCN features two key innovations: (1) a novel hyperbolic feature transformation layer that enables fast and accurate linear mappings, and (2) techniques such as hyperbolic residual connections and regularization for both weights and features, facilitated by an efficient hyperbolic midpoint method. Extensive experiments demonstrate that DeepHGCN achieves significant improvements in link prediction and node classification tasks compared to both Euclidean and shallow hyperbolic GCN variants.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
