Improving the Accuracy and Efficiency of MAP Inference for Markov Logic
This work addresses efficiency and accuracy issues in MAP inference for Statistical Relational Learning, which is incremental as it builds on existing methods like MaxWalkSAT and ILP.
The authors tackled the problem of slow and inaccurate MAP inference in Markov Logic by introducing Cutting Plane Inference (CPI), a meta algorithm that instantiates small parts of the network and solves them with conventional methods. They showed that CPI significantly speeds up both MaxWalkSAT and Integer Linear Programming while improving MaxWalkSAT's accuracy and maintaining exactness for ILP.
In this work we present Cutting Plane Inference (CPI), a Maximum A Posteriori (MAP) inference method for Statistical Relational Learning. Framed in terms of Markov Logic and inspired by the Cutting Plane Method, it can be seen as a meta algorithm that instantiates small parts of a large and complex Markov Network and then solves these using a conventional MAP method. We evaluate CPI on two tasks, Semantic Role Labelling and Joint Entity Resolution, while plugging in two different MAP inference methods: the current method of choice for MAP inference in Markov Logic, MaxWalkSAT, and Integer Linear Programming. We observe that when used with CPI both methods are significantly faster than when used alone. In addition, CPI improves the accuracy of MaxWalkSAT and maintains the exactness of Integer Linear Programming.