The reviewed record of science sign in
Pith

arxiv: 2406.12353 · v1 · pith:YUX4V66I · submitted 2024-06-18 · stat.ML · cs.LG

Top-Down Bayesian Posterior Sampling for Sum-Product Networks

Reviewed by Pithpith:YUX4V66Iopen to challenge →

classification stat.ML cs.LG
keywords spnscomplexitysamplingbayesiancomputationalconstraintsdistributionefficiently
0
0 comments X
read the original abstract

Sum-product networks (SPNs) are probabilistic models characterized by exact and fast evaluation of fundamental probabilistic operations. Its superior computational tractability has led to applications in many fields, such as machine learning with time constraints or accuracy requirements and real-time systems. The structural constraints of SPNs supporting fast inference, however, lead to increased learning-time complexity and can be an obstacle to building highly expressive SPNs. This study aimed to develop a Bayesian learning approach that can be efficiently implemented on large-scale SPNs. We derived a new full conditional probability of Gibbs sampling by marginalizing multiple random variables to expeditiously obtain the posterior distribution. The complexity analysis revealed that our sampling algorithm works efficiently even for the largest possible SPN. Furthermore, we proposed a hyperparameter tuning method that balances the diversity of the prior distribution and optimization efficiency in large-scale SPNs. Our method has improved learning-time complexity and demonstrated computational speed tens to more than one hundred times faster and superior predictive performance in numerical experiments on more than 20 datasets.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.