arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.15852cs.LGcs.AI
收藏

基于层次支撑图的下一代消息传递

Next Level Message-Passing with Hierarchical Support Graphs

ETH Zurich(苏黎世联邦理工学院)

Carlos Vonessen, Florian Grötschla, Roger Wattenhofer

AI总结:

针对MPNNs信息交换受限的问题,提出层次支撑图(HSG)通过递归粗化扩展虚拟节点概念,提供灵活框架增强信息流,并在多个数据集上取得最先进结果。

摘要:

消息传递神经网络(MPNNs)在图学习任务中被广泛使用,但存在局限性,例如在每轮消息传递期间信息交换仅局限于相邻节点,导致范围受限。为解决这些局限性,已有多种策略被提出,包括引入虚拟节点以促进全局信息交换。在本研究中,我们引入了层次支撑图(HSG),这是通过对原始图进行递归粗化而创建的虚拟节点概念的扩展。该方法提供了一个灵活的框架来增强图中的信息流,且不依赖于所使用的特定MPNN层。我们对HSGs进行了理论分析,研究了其经验性能,并证明HSGs能够超越其他使用虚拟节点增强的方法,在多个数据集上取得最先进的结果。

英文摘要:

Message-Passing Neural Networks (MPNNs) are extensively employed in graph learning tasks but suffer from limitations such as the restricted scope of information exchange, by being confined to neighboring nodes during each round of message passing. Various strategies have been proposed to address these limitations, including incorporating virtual nodes to facilitate global information exchange. In this study, we introduce the Hierarchical Support Graph (HSG), an extension of the virtual node concept created through recursive coarsening of the original graph. This approach provides a flexible framework for enhancing information flow in graphs, independent of the specific MPNN layers utilized. We present a theoretical analysis of HSGs, investigate their empirical performance, and demonstrate that HSGs can surpass other methods augmented with virtual nodes, achieving state-of-the-art results across multiple datasets.

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。

↑