arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.03548cs.LGstat.ML
收藏

图子神经网络与图神经网络的可移植性

Graphon Neural Networks and the Transferability of Graph Neural Networks

University of Pennsylvania(宾夕法尼亚大学);Dept. of Electrical and Systems Eng.(电气与系统工程系);Philadelphia, PA 19143(费城,PA 19143)

Luana Ruiz, Luiz F. O. Chamon, Alejandro Ribeiro

摘要:

图神经网络(GNN)依赖于图卷积来从网络数据中提取局部特征。这些图卷积使用所有节点共享的系数组合来自相邻节点的信息。由于这些系数是共享的,并且不依赖于图,因此可以设想使用相同的系数来定义另一个图上的GNN。这促使分析GNN在图中的可转移性。在本文中,我们引入图子NN作为GNN的极限对象,并证明了GNN的输出与其极限图子NN的输出之间的差的界。如果图卷积滤波器在图谱域中是带限的,则该界限随着节点数量的增加而消失。这一结果建立了GNN的可区分性和可转移性之间的权衡。

英文摘要:

Graph neural networks (GNNs) rely on graph convolutions to extract local features from network data. These graph convolutions combine information from adjacent nodes using coefficients that are shared across all nodes. Since these coefficients are shared and do not depend on the graph, one can envision using the same coefficients to define a GNN on another graph. This motivates analyzing the transferability of GNNs across graphs. In this paper we introduce graphon NNs as limit objects of GNNs and prove a bound on the difference between the output of a GNN and its limit graphon-NN. This bound vanishes with growing number of nodes if the graph convolutional filters are bandlimited in the graph spectral domain. This result establishes a tradeoff between discriminability and transferability of 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。

↑