Decision Sum-Product-Max Networks

Authors

  • Mazen Melibari University of Waterloo
  • Pascal Poupart University of Waterloo
  • Prashant Doshi University of Georgia

DOI:

https://doi.org/10.1609/aaai.v30i1.9957

Keywords:

Tractable Models, Sum-Product Networks, Decision Making Under Uncertainty

Abstract

Sum-Product Networks (SPNs) were recently proposed as a new class of probabilistic graphical models that guarantee tractable inference, even on models with high-treewidth. In this paper, we propose a new extension to SPNs, called Decision Sum-Product-Max Networks (Decision-SPMNs), that makes SPNs suitable for discrete multi-stage decision problems. We present an algorithm that solves Decision-SPMNs in a time that is linear in the size of the network. We also present algorithms to learn the parameters of the network from data.

AAAI 2016 Proceedings Cover

Downloads

Published

2016-03-05

How to Cite

Melibari, M., Poupart, P., & Doshi, P. (2016). Decision Sum-Product-Max Networks. Proceedings of the AAAI Conference on Artificial Intelligence, 30(1). https://doi.org/10.1609/aaai.v30i1.9957

Issue

Vol. 30 No. 1 (2016): Thirtieth AAAI Conference on Artificial Intelligence

Section

Student Abstracts and Posters