Priors on exchangeable directed graphs
arXiv:1510.08440 · doi:10.1214/16-EJS1185
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.
arXiv comment27 pages, 11 figures
References in corpus (15)
- Mixed membership stochastic blockmodels
- Matrix estimation by Universal Singular Value Thresholding
- Graph limits and exchangeable random graphs
- Sparse graphs using exchangeable random measures
- Sparse exchangeable graphs and their limits via graphon processes
- On exchangeable random variables and the statistics of large graphs and hypergraphs
- The Class of Random Graphs Arising from Exchangeable Random Measures
- An Infinite Latent Attribute Model for Network Data
- A Consistent Histogram Estimator for Exchangeable Graph Models
- Edge-exchangeable graphs and sparsity
- Nonparametric Multi-group Membership Model for Dynamic Networks
- Edge exchangeable models for network data
- Limits, Regularity and Removal for Finite Structures
- Multigraph limits, unbounded kernels, and Banach space decorated graphs
- Decomposition of tournament limits