Publication : Learning parameters for the sequence constraint from solutions
En cours de chargement...
Date
2016-08-23
Direction de publication
Direction de recherche
Titre de la revue
ISSN de la revue
Titre du volume
Éditeur
SpringerLink
Résumé
This paper studies the problem of learning parameters for global constraints such as Sequence from a small set of positive examples. The proposed technique computes the probability of observing a given constraint in a random solution. This probability is used to select the more likely constraint in a list of candidates. The learning method can be applied to both soft and hard constraints
Description
Revue
Principles and Practice of Constraint Programming, 405-420 (2016)
DOI
URL vers la version publiée
Mots-clés
Constraint Acquisition, Timetabling, Machine Learning, CSP, Solution Counting, Markov Chain, Soft Constraints, Global Constraints