LiU Electronic Press
Full-text not available in DiVA
Author:
Nyblom, Per (Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab) (Linköping University, The Institute of Technology)
Doherty, Patrick (Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab) (Linköping University, The Institute of Technology)
Title:
Towards Automatic Model Generation by Optimization
Department:
Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab
Linköping University, The Institute of Technology
Publication type:
Conference paper (Refereed)
Language:
English
In:
Proceedings of the Tenth Scandinavian Conference on Artificial Intelligence (SCAI)
Conference:
10th Scandinavian Conference on Artificial Intelligence (SCAI 2008), 26-28 May 2008, Stockholm, Sweden
Place of publ.: Amsterdam Publisher: IOS Press
Series:
Frontiers in Artificial Intelligence and Applications, ISSN 0922-6389; 173
Volume:
173
Pages:
114-123
Year of publ.:
2008
URI:
urn:nbn:se:liu:diva-58464
Permanent link:
http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-58464
ISBN:
978-1-58603-867-0, e-978-1-60750-335-4
ISI:
000273520700015
Subject category:
Engineering and Technology
Keywords(en) :
Automatic model generation
Abstract(en) :

The problem of automatically selecting simulation models for autonomous agents depending on their current intentions and beliefs is considered in this paper. The intended use of the models is for prediction, filtering, planning and other types of reasoning that can be performed with Simulation models. The parameters and model fragments of the resulting model are selected by formulating and solving a hybrid constrained optimization problem that captures the intuition of the preferred model when relevance information about the elements of the world being modelled is taken into consideration. A specialized version of the original optimization problem is developed that makes it possible to solve the continuous subproblem analytically in linear time. A practical model selection problem is discussed where the aim is to select suitable parameters and models for tracking dynamic objects. Experiments with randomly generated problem instances indicate that a hillclimbing search approach might be both efficient and provides reasonably good solutions compared to simulated annealing and hillclimbing with random restarts.

Available from:
2010-08-12
Created:
2010-08-11
Last updated:
2013-08-29
Statistics:
43 hits