Lifted Weight Learning of Markov Logic Networks Revisited
This work addresses efficiency in probabilistic relational models for AI researchers, but it is incremental as it builds on prior algorithms and focuses on a specific 2-variable constraint.
The authors tackled the problem of lifted weight learning for Markov logic networks, specifically for 2-variable cases, and developed an algorithm that runs in polynomial time relative to domain size, based on existing lifted-inference and maximum entropy distribution methods.
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.