arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.12843cs.LG
收藏

基于聚合增强的图表示学习

Graph Representation Learning via Aggregation Enhancement

Technion(以色列理工学院);Habana Labs(Habana实验室)

Maxim Fishman, Chaim Baskin, Evgenii Zheltonozhskii, Almog David, Ron Banner, Avi Mendelson

摘要:

图神经网络(GNN)已经成为处理图结构数据的强大工具,但在有效地聚合和传播层之间的信息方面仍然面临挑战,这限制了它们的性能。我们用核回归(KR)方法来解决这个问题,使用KR损失作为自监督设置中的主要损失或作为监督设置中的正则化项。与最先进的技术相比,我们在多个转导和归纳节点分类数据集的两种情况下都显示出了实质性的性能改进,特别是对于深度网络。与互信息(MI)相反,KR损失是凸的,并且在高维情况下易于估计,即使它间接地最大化其输入之间的MI。我们的工作突出了KR推进图表示学习领域和增强GNN性能的潜力。复制我们实验的代码可以在https://github.com/Anonymous1252022/KR_for_GNNs上找到。

英文摘要:

Graph neural networks (GNNs) have become a powerful tool for processing graph-structured data but still face challenges in effectively aggregating and propagating information between layers, which limits their performance. We tackle this problem with the kernel regression (KR) approach, using KR loss as the primary loss in self-supervised settings or as a regularization term in supervised settings. We show substantial performance improvements compared to state-of-the-art in both scenarios on multiple transductive and inductive node classification datasets, especially for deep networks. As opposed to mutual information (MI), KR loss is convex and easy to estimate in high-dimensional cases, even though it indirectly maximizes the MI between its inputs. Our work highlights the potential of KR to advance the field of graph representation learning and enhance the performance of GNNs. The code to reproduce our experiments is available at https://github.com/Anonymous1252022/KR_for_GNNs

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。

↑