Sum-product networks: A survey

Iago París, Raquel Sánchez-Cauce, Francisco Javier Díez
4/2/2020
cs.LGcs.AI

Official code: SPFlow/SPFlow (314 stars, Apache-2.0, last commit 2026-04-14). 1 more implementation below.

Abstract

A sum-product network (SPN) is a probabilistic model, based on a rooted acyclic directed graph, in which terminal nodes represent univariate probability distributions and non-terminal nodes represent convex combinations (weighted sums) and products of probability functions. They are closely related to probabilistic graphical models, in particular to Bayesian networks with multiple context-specific independencies. Their main advantage is the possibility of building tractable models from data, i.e., models that can perform several inference tasks in time proportional to the number of links in the graph. They are somewhat similar to neural networks and can address the same kinds of problems, such as image processing and natural language understanding. This paper offers a survey of SPNs, including their definition, the main algorithms for inference and learning from data, the main applications, a brief review of software libraries, and a comparison with related models

View on arXivView PDF

Code Implementations(2)

SPFlow/SPFlowOfficial100%
Apache-2.0

Sum Product Flow: An Easy and Extensible Library for Sum-Product Networks

31483HTML, PythonMar 20, 20186 months agoApache-2.0
deep-learninggraphical-modelsmachine-learningmixture-modelprobabilistic-circuits+3 more
arranger1044/awesome-spnOfficial100%

A structured list of resources about Sum-Product Networks (SPNs)

25439Jul 10, 20165 years ago
codeliteraturespnsum-product-networks

Similar Papers

Sum-Product-Set Networks: Deep Tractable Models for Tree-Structured Graphs

88% match2024

Tractable Probabilistic Graph Representation Learning with Graph-Induced Sum-Product Networks

87% match1 repo92023

GraphSPNs: Sum-Product Networks Benefit From Canonical Orderings

83% match1 repo22024

N-Parties Private Structure and Parameter Learning for Sum-Product Networks

79% match2025

χSPN: Characteristic Interventional Sum-Product Networks for Causal Inference in Hybrid Domains

79% match2024

Signed Diverse Multiplex Networks: Clustering and Inference

79% match2024

Cite this paper

@article{pars2020sumproduct,
  title  = {Sum-product networks: A survey},
  author = {Iago París and Raquel Sánchez-Cauce and Francisco Javier Díez},
  year   = {2020},
  eprint = {2004.01167},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2004.01167}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2004.01167.svg)](https://codeofpaper.com/paper/2004.01167)

Discussion