1. A Central Limit Theorem for an Omnibus Embedding of Multiple Random Dot Product Graphs
- Author
-
Keith Levin, Minh Tang, Carey E. Priebe, Avanti Athreya, and Vince Lyzinski
- Subjects
Power graph analysis ,Discrete mathematics ,Computer science ,Graph embedding ,Euclidean space ,Dot product ,01 natural sciences ,Graph ,Vertex (geometry) ,010104 statistics & probability ,03 medical and health sciences ,0302 clinical medicine ,Embedding ,0101 mathematics ,030217 neurology & neurosurgery ,MathematicsofComputing_DISCRETEMATHEMATICS ,Central limit theorem - Abstract
Performing statistical inference on collections of graphs is of import to many disciplines. Graph embedding, in which the vertices of a graph are mapped to vectors in a low-dimensional Euclidean space, has gained traction as a basic tool for graph analysis. Here we describe an omnibus embedding in which multiple graphs on the same vertex set are jointly embedded into a single space with a distinct representation for each graph. We prove a central limit theorem for this omnibus embedding and show that simultaneous embedding into a common space allows comparison of graphs without the need to perform pairwise alignments of graph embeddings. Experimental results demonstrate that the omnibus embedding improves upon existing methods. (Please note: per the D3M workshop organizers' request, this is an extended abstract of a paper that has already been posted to arXiv at https://arxiv.org/abs/1705.09355)
- Published
- 2017