DeepAI AI Chat
Log In Sign Up

Bayesian structure learning using dynamic programming and MCMC

by   Daniel Eaton, et al.

MCMC methods for sampling from the space of DAGs can mix poorly due to the local nature of the proposals that are commonly used. It has been shown that sampling from the space of node orders yields better results [FK03, EW06]. Recently, Koivisto and Sood showed how one can analytically marginalize over orders using dynamic programming (DP) [KS04, Koi06]. Their method computes the exact marginal posterior edge probabilities, thus avoiding the need for MCMC. Unfortunately, there are four drawbacks to the DP technique: it can only use modular priors, it can only compute posteriors over modular features, it is difficult to compute a predictive density, and it takes exponential time and space. We show how to overcome the first three of these problems by using the DP algorithm as a proposal distribution for MCMC in DAG space. We show that this hybrid technique converges to the posterior faster than other methods, resulting in more accurate structure learning and higher predictive likelihoods on test data.


A Parallel Algorithm for Exact Bayesian Structure Discovery in Bayesian Networks

Exact Bayesian structure discovery in Bayesian networks requires exponen...

Differentially Private Hamiltonian Monte Carlo

Markov chain Monte Carlo (MCMC) algorithms have long been the main workh...

Scaling Structured Inference with Randomization

Deep discrete structured models have seen considerable progress recently...

A Novel Dynamic Programming Approach to the Train Marshalling Problem

Train marshalling is the process of reordering the railcars of a train i...

Tree-Guided MCMC Inference for Normalized Random Measure Mixture Models

Normalized random measures (NRMs) provide a broad class of discrete rand...

Posterior inference unchained with EL_2O

Statistical inference of analytically non-tractable posteriors is a diff...

Dynamic Maintenance of Monotone Dynamic Programs and Applications

Dynamic programming (DP) is one of the fundamental paradigms in algorith...