Credal Sum-Product Networks

Denis Deratani Maua, Fabio Gagli Cozman, Diarmaid Conaty, Casio Polpo de Campos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

294 Downloads (Pure)

Abstract

Sum-product networks are a relatively new and increasingly popular class of (precise) probabilistic graphical models that allow for marginal inference with polynomial effort. As with other probabilistic models, sum-product networks are often learned from data and used to perform classification. Hence, their results are prone to be unreliable and overconfident. In this work, we develop credal sum-product networks, an imprecise extension of sum-product networks. We present algorithms and complexity results for common inference tasks. We apply our algorithms on realistic classification task using images of digits and show that credal sum-product networks obtained by a perturbation of the parameters of learned sum-product networks are able to distinguish between reliable and unreliable classifications with high accuracy.
Original languageEnglish
Title of host publicationISIPTA'17: Proceedings of the Tenth International Symposium on Imprecise Probability: Theories and Applications
Pages205-216
Number of pages12
Publication statusPublished - 14 Jul 2017
EventTenth International Symposium on Imprecise Probability: Theories and Applications (ISIPTA ’17) - Lugano, Switzerland
Duration: 10 Jul 201714 Jul 2017
http://conference.researchbib.com/view/event/64182

Publication series

NameProceedings of Machine Learning Research
Volume62
ISSN (Print)1938-7228

Conference

ConferenceTenth International Symposium on Imprecise Probability: Theories and Applications (ISIPTA ’17)
Abbreviated titleISIPTA 2017
CountrySwitzerland
CityLugano
Period10/07/201714/07/2017
Internet address

Fingerprint

Dive into the research topics of 'Credal Sum-Product Networks'. Together they form a unique fingerprint.

Cite this