Priors on exchangeable directed graphs

DOI10.1214/16-EJS1185zbMATH Open1355.60041arXiv1510.08440OpenAlexW2256584667MaRDI QIDQ502818FDOQ502818

Authors: Diana Cai, Nathanael Ackerman, Cameron E. Freer 

Publication date: 11 January 2017

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: Directed graphs occur throughout statistical modeling of networks, and exchangeability is a natural assumption when the ordering of vertices does not matter. There is a deep structural theory for exchangeable undirected graphs, which extends to the directed case via measurable objects known as digraphons. Using digraphons, we first show how to construct models for exchangeable directed graphs, including special cases such as tournaments, linear orderings, directed acyclic graphs, and partial orderings. We then show how to construct priors on digraphons via the infinite relational digraphon model (di-IRM), a new Bayesian nonparametric block model for exchangeable directed graphs, and demonstrate inference on synthetic data.

Full work available at URL: https://arxiv.org/abs/1510.08440

Recommendations

• A generalization of hierarchical exchangeability on trees to directed acyclic graphs
• On exchangeability in network models
• Bayesian Models for Directed Graphs
• Prior distributions for Gaussian graphical models: a comparison between the directed and undirected case
• Network representation using graph root distributions

zbMATH Keywords

graphonnetwork modeldirected graphblock modeldigraphonexchangeable graphnonparametric prior

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Directed graphs (digraphs), tournaments (05C20) Exchangeability for stochastic processes (60G09)

This page was built for publication: Priors on exchangeable directed graphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q502818)