Classifier-based constraint acquisition
Publication Type:
Refereed Original Article
Abstract:
Modeling a combinatorial problem is a hard and error-prone task requiring significant expertise. Constraint acquisition methods attempt to automate this process by learning constraints from examples of solutions and (usually) non-solutions. Active methods query an oracle while passive methods do not. We propose a known but not widely-used application of machine learning to constraint acquisition: training a classifier to discriminate between solutions and non-solutions, then deriving a constraint model from the trained classifier. We discuss a wide range of possible new acquisition methods with useful properties inherited from classifiers. We also show the potential of this approach using a Naive Bayes classifier, obtaining a new passive acquisition algorithm that is considerably faster than existing methods, scalable to large constraint sets, and robust under errors.
Digital Object Identifer (DOI):
10.1007/s10472-021-09736-4
Publication Status:
Published
Date Accepted for Publication:
Monday, 1 March, 2021
Publication Date:
17/04/2021
Journal:
Annals of Mathematics and Artificial Intelligence
Research Group:
Institution:
National University of Ireland, Cork (UCC)
Open access repository:
Yes