Summary: Improving the Efficiency of Reasoning Through
Structure­Based Reformulation #
Eyal Amir 1 and Sheila McIlraith 2
1 Department of Computer Science, Stanford University, Stanford, CA 94305,
eyal.amir@cs.stanford.edu
2 Knowledge Systems Lab, Department of Computer Science, Stanford University,
Stanford, CA 94305, sheila.mcilraith@cs.stanford.edu
Abstract. We investigate the possibility of improving the efficiency of reasoning
through structure­based partitioning of logical theories, combined with partition­
based logical reasoning strategies. To this end, we provide algorithms for reason­
ing with partitions of axioms in first­order and propositional logic. We analyze
the computational benefit of our algorithms and detect those parameters of a par­
titioning that influence the efficiency of computation. These parameters are the
number of symbols shared by a pair of partitions, the size of each partition, and
the topology of the partitioning. Finally, we provide a greedy algorithm that au­
tomatically reformulates a given theory into partitions, exploiting the parameters
that influence the efficiency of computation.
1 Introduction
There is growing interest in building large knowledge bases (KBs) of everyday knowl­
edge about the world, teamed with theorem provers to perform inference. Three such

Source: Amir, Eyal - Department of Computer Science, University of Illinois at Urbana-Champaign

Collections: Computer Technologies and Information Sciences