Asymmetric Graph Representation Learning
Abstract
Despite the enormous success of graph neural networks (GNNs), most existing GNNs can only be applicable to undirected graphs where relationships among connected nodes are twoway symmetric (i.e., information can be passed back and forth). However, there is a vast amount of applications where the information flow is asymmetric, leading to directed graphs where information can only be passed in one direction. For example, a directed edge indicates that the information can only be conveyed forwardly from the start node to the end node, but not backwardly. To accommodate such an asymmetric structure of directed graphs within the framework of GNNs, we propose a simple yet remarkably effective framework for directed graph analysis to incorporate such oneway information passing. We define an incoming embedding and an outgoing embedding for each node to model its sending and receiving features respectively. We further develop two steps in our directed GNN model with the first one to aggregate/update the incoming features of nodes and the second one to aggregate/update the outgoing features. By imposing the two roles for each node, the likelihood of a directed edge can be calculated based on the outgoing embedding of the start node and the incoming embedding of the end node. The loglikelihood of all edges plays a natural role of regularization for the proposed model, which can alleviate the oversmoothing problem of the deep GNNs. Extensive experiments on multiple realworld directed graphs demonstrate outstanding performances of the proposed model in both nodelevel and graphlevel tasks.
 Publication:

arXiv eprints
 Pub Date:
 October 2021
 DOI:
 10.48550/arXiv.2110.07436
 arXiv:
 arXiv:2110.07436
 Bibcode:
 2021arXiv211007436T
 Keywords:

 Computer Science  Machine Learning