Low-dimensional statistical manifold embedding of directed graphs
This work addresses the challenge of representing directed graphs for applications in network analysis, though it appears incremental as it builds on existing embedding techniques.
The authors tackled the problem of embedding directed graphs into statistical manifolds by minimizing pairwise relative entropy and graph geodesics, resulting in an unsupervised method that outperforms existing models on various metrics and better preserves global geodesic information.
We propose a novel node embedding of directed graphs to statistical manifolds, which is based on a global minimization of pairwise relative entropy and graph geodesics in a non-linear way. Each node is encoded with a probability density function over a measurable space. Furthermore, we analyze the connection between the geometrical properties of such embedding and their efficient learning procedure. Extensive experiments show that our proposed embedding is better in preserving the global geodesic information of graphs, as well as outperforming existing embedding models on directed graphs in a variety of evaluation metrics, in an unsupervised setting.