Skip to content
conftrace_
2019 AISTATS AISTATS 2019

Lifted Weight Learning of Markov Logic Networks Revisited

Abstract

We study lifted weight learning of Markov logic networks. We show that there is an algorithm for maximum-likelihood learning of 2-variable Markov logic networks which runs in time polynomial in the domain size. Our results are based on existing lifted-inference algorithms and recent algorithmic results on computing maximum entropy distributions.

Authors

Ondrej Kuzelka, Vyacheslav Kungurtsev

Topics

Artificial Intelligence > Bayesian & Probabilistic > Probabilistic Modeling Machine Learning > Optimization & Theory > Statistical Learning

Keywords

maximum likelihood estimation lifted inference markov logic network maximum entropy distribution
Download PDF

More from AISTATS 2019

All AISTATS 2019 papers →
Truncated Back-propagation for Bilevel Optimization
Empirical Risk Minimization and Stochastic Gradient Descent for Relational Data
Variable selection for Gaussian processes via sensitivity analysis of the posterior predictive distribution
Causal Discovery in the Presence of Missing Data
Learning Tree Structures from Noisy Data
Active multiple matrix completion with adaptive confidence sets