A k-elitist max-min ant system approach to cost-based abduction

Abstract

Abduction is the process of proceeding from data describing a set of observations or events, to a set of hypotheses which best explains or accounts for the data. Cost-based abduction (CBA) is a formalism in which evidence to be explained is treated as a goal to be proven, proofs have costs based on how much needs to be assumed to complete the proof, and the set of assumptions needed to complete the least-cost proof are taken as the best explanation for the given evidence. We apply a k-elitist variation on the max-min ant system (MMAS) to CBA, in which the k-best ants are allowed to update the global pheromone trace array in every iteration; in the original MMAS, only the single best ant updates the trace array (thus, it can be considered 1-elitist). Applying our technique to several large CBA instances, we find that our k-elitist approach, with k varying in our experiments from 1 to 15, returns lower-cost proofs on average than the original MMAS. A test of statistical significance is used to verify that the differences in performance are statistically significant.