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.)

Speeding up parameter and rule learning for acyclic probabilistic logic programs

Full text
Author(s):
de Faria, Francisco H. O. Vieira [1] ; Gusmao, Arthur Colombini [1] ; De Bona, Glauber [1] ; Maua, Denis Deratani [2] ; Cozman, Fabio Gagliardi [1]
Total Authors: 5
Affiliation:
[1] Univ Sao Paulo, Escola Politecn, Sao Paulo - Brazil
[2] Univ Sao Paulo, Inst Matemat & Estat, Sao Paulo - Brazil
Total Affiliations: 2
Document type: Journal article
Source: INTERNATIONAL JOURNAL OF APPROXIMATE REASONING; v. 106, p. 32-50, MAR 2019.
Web of Science Citations: 0
Abstract

This paper introduces techniques that speed-up parameter and rule learning for acyclic probabilistic logic programs. We focus on maximum likelihood estimation of parameters, and show that significant improvements can be obtained by efficiently handling probabilistic rules. We then move to structure learning, where we learn sets of rules, by introducing an algorithm that greatly simplifies exact score-based learning. Experiments demonstrate that our methods can produce orders of magnitude speed-ups over the state-of-art in parameter and rule learning. (C) 2018 Elsevier Inc. All rights reserved. (AU)

FAPESP's process: 17/19007-6 - Inference and learning algorithms for probabilistic logic programming
Grantee:Arthur Colombini Gusmão
Support Opportunities: Scholarships in Brazil - Master
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
FAPESP's process: 16/25928-4 - Probabilistic extensions for fragments of first-order logic
Grantee:Glauber de Bona
Support Opportunities: Scholarships in Brazil - Post-Doctorate
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: 15/21880-4 - PROVERBS -- PRobabilistic OVERconstrained Boolean Systems: reasoning tools and applications
Grantee:Marcelo Finger
Support Opportunities: Regular Research Grants