Advanced search
Start date
Betweenand
Related content
(Reference retrieved automatically from Web of Science through information on FAPESP grant and its corresponding number as mentioned in the publication by the authors.)

Robustifying sum-product networks

Full text
Author(s):
Maua, Denis Deratani [1] ; Conaty, Diarmaid [2] ; Cozman, Fabio Gagliardi [3] ; Poppenhaeger, Katja [4] ; de Campos, Cassio Polpo [2, 5]
Total Authors: 5
Affiliation:
[1] Univ Sao Paulo, Inst Math & Stat, Sao Paulo - Brazil
[2] Queens Univ Belfast, Ctr Data Sci & Scalable Comp, Belfast, Antrim - North Ireland
[3] Univ Sao Paulo, Escola Politecn, Sao Paulo - Brazil
[4] Queens Univ Belfast, Astrophys Res Ctr, Belfast, Antrim - North Ireland
[5] Univ Utrecht, Dept Informat & Comp Sci, Utrecht - Netherlands
Total Affiliations: 5
Document type: Journal article
Source: INTERNATIONAL JOURNAL OF APPROXIMATE REASONING; v. 101, n. SI, p. 163-180, OCT 2018.
Web of Science Citations: 0
Abstract

Sum-product networks are a relatively new and increasingly popular family of probabilistic graphical models that allow for marginal inference with polynomial effort. They have been shown to achieve state-of-the-art performance in several tasks involving density estimation. Sum-product networks are typically learned from data; as such, inferences produced with them are prone to be unreliable and overconfident when data is scarce. In this work, we develop the credal sum-product networks, a generalization of sum-product networks that uses set-valued parameters. We present algorithms and complexity results for common inference tasks with this class of models. We also present an approach for assessing the reliability of classifications made with sum-product networks. We apply this approach on benchmark classification tasks as well as a new application in predicting the age of stars. Our experiments show that the use of credal sum-product networks allow us to distinguish between reliable and unreliable classifications with higher accuracy than standard approaches based on (precise) probability values. (C) 2018 Elsevier Inc. All rights reserved. (AU)

FAPESP's process: 16/01055-1 - Learning of Tractable Probabilistic Models with Application to Multilabel Classification
Grantee:Denis Deratani Mauá
Support Opportunities: Regular Research Grants
FAPESP's process: 16/18841-0 - Inference and learning algorithms for probabilistic logic programming
Grantee:Fabio Gagliardi Cozman
Support Opportunities: Research Grants - Research Partnership for Technological Innovation - PITE