Exploiting causal independence in Markov logic networks: combining undirected and directed models

Authors:
Sriraam Natarajan;Tushar Khot;Daniel Lowd;Prasad Tadepalli;Kristian Kersting;Jude Shavlik
Affiliations:
University of Wisconsin-Madison;University of Wisconsin-Madison;University of Oregon;Oregon State University;Fraunhofer IAIS;University of Wisconsin-Madison
Venue:
ECML PKDD'10 Proceedings of the 2010 European conference on Machine learning and knowledge discovery in databases: Part II
Year:
2010

Citing 11
Cited 0

Learning Probabilistic Models of Relational Structure

ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
PRL: A probabilistic relational language

Machine Learning
Parameter learning for relational Bayesian networks

Proceedings of the 24th international conference on Machine learning
Introduction to Statistical Relational Learning (Adaptive Computation and Machine Learning)

Introduction to Statistical Relational Learning (Adaptive Computation and Machine Learning)
Learning first-order probabilistic models with combining rules

Annals of Mathematics and Artificial Intelligence
Learning probabilities for noisy first-order rules

IJCAI'97 Proceedings of the Fifteenth international joint conference on Artifical intelligence - Volume 2
Exploiting causal independence in Bayesian network inference

Journal of Artificial Intelligence Research
Model-theoretic expressivity analysis

Probabilistic inductive logic programming
Markov Logic: An Interface Layer for Artificial Intelligence

Markov Logic: An Interface Layer for Artificial Intelligence
Relational bayesian networks

UAI'97 Proceedings of the Thirteenth conference on Uncertainty in artificial intelligence
Logical bayesian networks and their relation to other probabilistic logical models

ILP'05 Proceedings of the 15th international conference on Inductive Logic Programming

Quantified Score

Hi-index	0.01

Visualization

Abstract

A new method is proposed for compiling causal independencies into Markov logic networks (MLNs). An MLN can be viewed as compactly representing a factorization of a joint probability into the product of a set of factors guided by logical formulas. We present a notion of causal independence that enables one to further factorize the factors into a combination of even smaller factors and consequently obtain a finer-grain factorization of the joint probability. The causal independence lets us specify the factor in terms of weighted, directed clauses and operators, such as "or", "sum" or "max", on the contribution of the variables involved in the factors, hence combining both undirected and directed knowledge. Our experimental evaluations shows that making use of the finer-grain factorization provided by causal independence can improve quality of parameter learning in MLNs.