Estimating Dependency Structure as a Hidden Variable
Author(s)Meila, Marina; Jordan, Michael I.; Morris, Quaid
MetadataShow full item record
This paper introduces a probability model, the mixture of trees that can account for sparse, dynamically changing dependence relationships. We present a family of efficient algorithms that use EMand the Minimum Spanning Tree algorithm to find the ML and MAP mixtureof trees for a variety of priors, including the Dirichlet and the MDL priors.