Brandon Mayer, Anton Tsitsulin, Hendrik Fichtenberger, Jonathan Halcrow, Bryan Perozzi
TL;DR图嵌入是网络理解的常用方法，本论文提出了一种基于 Tensor Processing Units 的高性能图嵌入架构，能够解决大规模图的嵌入问题，并通过真实和合成的大规模数据集验证了嵌入空间的质量。
Abstract
graphs are a representation of structured data that captures the relationships between sets of objects. With the ubiquity of available network data, there is increasing industrial and academic need to quickly analyze gr