Abstract
The dominant approach for learning Bayesian networks from data is based on the use ofa scoring metric, that evaluates the fitness of any given candidate network to the data, and a search procedure, that explores the space ofp ossible solutions. The most efficient methods used in this context are (Iterated) Local Search algorithms. These methods use a predefined neighborhood structure that defines the feasible elementary modifications (local changes) that can be applied to a given solution in order to get another, potentially better solution. Ifthe search space is the set of directed acyclic graphs (dags), the usual choices for local changes are arc addition, arc deletion and arc reversal. In this paper we propose a new definition ofneigh borhood in the dag space, which uses a modified operator for arc reversal. The motivation for this new operator is the observation that local search algorithms experience problems when some arcs are wrongly oriented. We exemplify the general usefulness of our proposal by means ofa set ofexp eriments with different metrics and different local search methods, including Hill-Climbing and Greedy Randomized Adaptive Search Procedure (GRASP), as well as using several domain problems.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
S. Acid and L. M. de Campos. A hybrid methodology for learning belief networks: Benedict. International Journal of Approximate Reasoning, 27(3):235–262, 2001.
J. Binder, D. Koller, S. Russell, and K. Kanazawa. Adaptive probabilistic networks with hidden variables. Machine Learning, 29(2):213–244, 1997.
I. A. Beinlich, H. J. Suermondt, R. M. Chavez, and G. F. Cooper. The ALARM monitoring system: A case study with two probabilistic inference techniques for beliefnet works. In Proceedings of the Second European Conference on Artificial Intelligence in Medicine, 247–256, 1989.
R. R. Bouckaert. Bayesian BeliefNet works: From Construction to Inference. PhD. Thesis, University ofUtrec ht, 1995.
W. Buntine. Theory refinement of Ba yesian networks. In Proceedings of the Seventh Conference on Uncertaintyin Artificial Intelligence, 52–60, 1991.
W. Buntine. A guide to the literature on learning probabilistic networks from data. IEEE Transactions on Knowledge and Data Engineering, 8:195–210, 1996.
D. M. Chickering, D. Geiger, and D. Heckerman. Learning Bayesian networks is NP-Complete. In D. Fisher and H. Lenz, Eds., Learning from Data: Artificial Intelligence and Statistics V, Springer-Verlag, 121–130, 1996.
G. F. Cooper and E. Herskovits. A Bayesian method for the induction of probabilistic networks from data. Machine Learning, 9(4):309–348, 1992.
D. Dash and M. Druzdel. A hybrid anytime algorithm for the construction of causal models from sparse data. In Proceedings of the Fifteenth Conference on Uncertaintyin Artificial Intelligence, 142–149, 1999.
L. M. de Campos and J. F. Huete. A new approach for learning belief networks using independence criteria. International Journal of Approximate Reasoning, 24:11–37, 2000.
L. M. de Campos and J. M. Puerta. Stochastic local search algorithms for learning beliefnet works: Searching in the space oforderings. Lecture Notes in Artificial Intelligence, 2143:228–239, 2001.
T. A. Feo and M. G. C. Resende. Greedy randomized adaptive search procedures. Journal of Global Optimization, 6:109–133, 1995.
D. Heckerman, D. Geiger, and D. M. Chickering. Learning Bayesian networks: The combination ofkno wledge and statistical data. Machine Learning, 20:197–244, 1995.
F. V. Jensen. An Introduction to Bayesian Networks. UCL Press, 1996.
T. Kocka and R. Castelo. Improved learning ofBa yesian networks. In Proceedings of the Seventeenth Conference on Uncertaintyin Artificial Intelligence, 269–276, 2001.
P. Larrañaga, M. Poza, Y. Yurramendi, R. Murga, and C. Kuijpers. Structure learning of Ba yesian networks by genetic algorithms: A performance analysis of control parameters. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(9):912–926, 1996.
W. Lam and F. Bacchus. Learning Bayesian beliefnet works. An approach based on the MDL principle. Computational Intelligence, 10(4):269–293, 1994.
J. Pearl. Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, San Mateo, 1988.
J. Pearl and T. S. Verma. Equivalence and synthesis ofcausal models. In Proceedings of the Sixth Conference on Uncertaintyin Artificial Intelligence, 220–227, 1990.
M. G. C. Resende and C. C. Ribeiro. Greedy randomized adaptive search procedures. In F. Glover and G. Kochenberger, Eds., State of the Art Handbook in Metaheuristics, Kluwer. To appear.
P. Spirtes, C. Glymour, and R. Scheines. Causation, Prediction, and Search. Lecture Notes in Statistics 81, Springer Verlag, 1993.
M. Singh and M. Valtorta. Construction of Ba yesian network structures from data: A brief survey and an efficient algorithm. International Journal of Approximate Reasoning, 12:111–131, 1995.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Miguel de Campos, L., Manuel Fernández-Luna, J., Miguel Puerta, J. (2002). Local Search Methods for Learning Bayesian Networks Using a Modified Neighborhood in the Space of DAGs. In: Garijo, F.J., Riquelme, J.C., Toro, M. (eds) Advances in Artificial Intelligence — IBERAMIA 2002. IBERAMIA 2002. Lecture Notes in Computer Science(), vol 2527. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-36131-6_19
Download citation
DOI: https://doi.org/10.1007/3-540-36131-6_19
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-00131-7
Online ISBN: 978-3-540-36131-2
eBook Packages: Springer Book Archive