针对超图结构,本文提出了一种Hyper-SAGNN图神经网络,该网络可以用于处理异构超图,此处的异构性体现在超图中的hyperedges具有不同的size,并通过Self-attention机制来aggregate超边连接的节点。所设计的Hyper-SAGNN针对的应用为hyperedge prediction。 Hyperedge prediction problem (超边预测) 给定一个tuple(v1,v2,...,...
Hyper-SAGNN: a self-attention based graph neural network for hypergraphs,ICLR(2020)Ruochi Zhang,Yuesong Zou,Jian Ma 这篇文章针对的是hypergraph,并且使用的数据是scHi-C(single cell Hi-C),两种感觉都不是经常能见到的名词凑一起了。 本文处理的对象是Hi-C数据集,但在实验里也给出了其他可供参考的数据...
Hyper-SAGNN: a self-attention based graph neural network for hypergraphsRuochi ZhangYuesong ZouJian MaInternational Conference on Learning Representations
Here we develop a new self-attention based graph neural network called Hyper-SAGNN applicable to homogeneous and heterogeneous hypergraphs with variable hyperedge sizes. We perform extensive evaluations on multiple datasets, including four benchmark network datasets and two single-cell Hi-C datasets in...