Chapter

Symbolic and Quantitative Approaches to Reasoning with Uncertainty

Volume 6717 of the series Lecture Notes in Computer Science pp 110-121

Using Four Cost Measures to Determine Arc Reversal Orderings

  • Cory J. ButzAffiliated withDepartment of Computer Science, University of Regina
  • , Anders L. MadsenAffiliated withHUGIN EXPERT A/S
  • , Kevin WilliamsAffiliated withDepartment of Computer Science, University of Regina

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Four cost measures s 1, s 2, s 3, s 4 were recently studied for sorting the operations in Lazy propagation with arc reversal (LPAR), a join tree propagation approach to Bayesian network inference. It has been suggested to use s 1 with LPAR, since there is an effectiveness ranking, say s 1, s 2, s 3, s 4, when applied in isolation. In this paper, we also suggest to use s 1 with LPAR, but to use s 2 to break s 1 ties, s 3 to break s 2 ties, and s 4 to break s 3 ties. Experimental results show that sometimes there is a noticeable gain to be made.

Keywords

Bayesian networks arc reversal join tree propagation