Learning directed relational models with recursive dependencies

Schulte, Oliver, Khosravi, Hassan and Man, Tong (2012) Learning directed relational models with recursive dependencies. Machine Learning, 89 3: 299-316. doi:10.1007/s10994-012-5308-5


Author Schulte, Oliver
Khosravi, Hassan
Man, Tong
Title Learning directed relational models with recursive dependencies
Journal name Machine Learning   Check publisher's open access policy
ISSN 0885-6125
1573-0565
Publication date 2012-12-01
Sub-type Article (original research)
DOI 10.1007/s10994-012-5308-5
Open Access Status Not yet assessed
Volume 89
Issue 3
Start page 299
End page 316
Total pages 18
Place of publication New York, United States
Publisher Springer
Language eng
Abstract Recently, there has been an increasing interest in generative models that represent probabilistic patterns over both links and attributes. A common characteristic of relational data is that the value of a predicate often depends on values of the same predicate for related entities. For directed graphical models, such recursive dependencies lead to cycles, which violates the acyclicity constraint of Bayes nets. In this paper we present a new approach to learning directed relational models which utilizes two key concepts: a pseudo likelihood measure that is well defined for recursive dependencies, and the notion of stratification from logic programming. An issue for modelling recursive dependencies with Bayes nets are redundant edges that increase the complexity of learning. We propose a new normal form format that removes the redundancy, and prove that assuming stratification, the normal form constraints involve no loss of modelling power. Empirical evaluation compares our approach to learning recursive dependencies with undirected models (Markov Logic Networks). The Bayes net approach is orders of magnitude faster, and learns more recursive dependencies, which lead to more accurate predictions.
Keyword Autocorrelation
Bayesian networks
Recursive dependencies
Statistical relational learning
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status Non-UQ

Document type: Journal Article
Sub-type: Article (original research)
Collection: Institute for Teaching and Learning Innovation Publications
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 1 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 3 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Thu, 15 Sep 2016, 02:24:12 EST by Hassan Khosravi on behalf of Learning and Research Services (UQ Library)