COPT: Coordinated Optimal Transport on Graphs
NIPS 2020(2020)
摘要
We introduce COPT, a novel distance metric between graphs defined via an optimization routine, computing a coordinated pair of optimal transport maps simultaneously. This is an unsupervised way to learn general-purpose graph representations, it can be used for both graph sketching and graph comparison. COPT involves simultaneously optimizing dual transport plans, one between the vertices of two graphs, and another between graph signal probability distributions. We show both theoretically and empirically that our method preserves important global structural information on graphs, in particular spectral information, making it well-suited for tasks on graphs including retrieval, classification, summarization, and visualization.
更多查看译文
关键词
coordinated optimal transport,optimal transport,graphs
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络