arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2002.03575cs.LGcs.GRstat.ML
收藏

具有邻域交互作用的双线性图神经网络

Bilinear Graph Neural Network with Neighbor Interactions

University of Science and Technology of China(中国科学技术大学);National University of Singapore(新加坡国立大学);Beijing Kuaishou Technology Co., Ltd.(北京快手科技有限公司);University of Electronic Science and Technology of China(电子科技大学)

Hongmin Zhu, Fuli Feng, Xiangnan He, Xiang Wang, Yan Li, Kai Zheng, Yongdong Zhang

摘要:

图神经网络(GNN)是一种强大的模型,可以学习表示并对图数据进行预测。GNN上的现有工作在很大程度上将图卷积定义为连接节点的特征的加权和,以形成目标节点的表示。然而,加权和的操作假设邻居节点是彼此独立的,并且忽略了它们之间可能的相互作用。当存在这种相互作用时,例如两个邻居节点的共同出现是目标节点特征的强烈信号,现有的GNN模型可能无法捕获信号。在这项工作中,我们认为在GNN的邻居节点之间的相互作用建模的重要性。我们提出了一个新的图卷积算子,它增加了加权和与相邻节点的表示的成对相互作用。我们将这个框架称为双线性图神经网络(BGNN),它通过相邻节点之间的双线性交互来提高GNN的表示能力。特别是,我们指定两个BGNN模型命名为BGCN和BGAT,分别基于著名的GCN和GAT。在三个公开的半监督节点分类基准上的实验结果验证了BGNN的有效性-- BGCN(BGAT)在分类精度上优于GCN(GAT)1.6%(1.5%)。代码可在https://github.com/zhuhm1996/bgnn上获得。

英文摘要:

Graph Neural Network (GNN) is a powerful model to learn representations and make predictions on graph data. Existing efforts on GNN have largely defined the graph convolution as a weighted sum of the features of the connected nodes to form the representation of the target node. Nevertheless, the operation of weighted sum assumes the neighbor nodes are independent of each other, and ignores the possible interactions between them. When such interactions exist, such as the co-occurrence of two neighbor nodes is a strong signal of the target node's characteristics, existing GNN models may fail to capture the signal. In this work, we argue the importance of modeling the interactions between neighbor nodes in GNN. We propose a new graph convolution operator, which augments the weighted sum with pairwise interactions of the representations of neighbor nodes. We term this framework as Bilinear Graph Neural Network (BGNN), which improves GNN representation ability with bilinear interactions between neighbor nodes. In particular, we specify two BGNN models named BGCN and BGAT, based on the well-known GCN and GAT, respectively. Empirical results on three public benchmarks of semi-supervised node classification verify the effectiveness of BGNN -- BGCN (BGAT) outperforms GCN (GAT) by 1.6% (1.5%) in classification accuracy.Codes are available at: https://github.com/zhuhm1996/bgnn.

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。

↑