Abstract
This paper considers an abstract framework for expressing approximate inference algorithms in valuation based systems. It will provide a definition of a ‘more informative’ binary relation between representations of information as well as the basic properties of a divergence measure. The approach is illustrated with the cases of probabilistic reasoning (computation of marginal probabilities and most probable explanation) and with inference problems in propositional logic. Examples of divergence measures satisfying the basic properties will be given for these problems. Finally, we will formulate in an abstract way the mean field variational approach and the iterative belief propagation algorithm.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Shafer, G., Shenoy, P.: Local computation in hypertrees. Working Paper N. 201, School of Business, University of Kansas, Lawrence (1988)
Shenoy, P., Shafer, G.: Axioms for probability and belief-function propagation. In: Shachter, et al. (ed.) Uncertainty in Artificial Intelligence, vol. 4, pp. 169–198. Elsevier, Amsterdam (1990)
Cano, J., Delgado, M., Moral, S.: An axiomatic system for the propagation of uncertainty in directed acyclic networks. Int. J. Approx. Reasoning 8, 253–280 (1993)
Kohlas, J.: Information Algebras: Generic Structures for Inference. Springer, London (2012). https://doi.org/10.1007/978-1-4471-0009-6
Kohlas, J., Moral, S.: Propositional information systems. Working Paper N. 96–01, Institute of Informatics, University of Fribourg (1996)
Hernández, L., Moral, S.: Mixing exact and importance sampling propagation algorithms in dependence graphs. Int. J. Intell. Syst. 12, 553–576 (1997)
Pearl, J.: Evidential reasoning using stochastic simulation of causal models. Artif. Intell. 32, 247–257 (1987)
Hernández, L., Moral, S., Salmerón, A.: Importance sampling algorithms in belief networks based on aproximate computation. In: Proceedings of Information Processing and Management of Uncertainty in Knowledge-Based Systems Conference (IPMU 1996), vol. 2, Granada, pp. 859–864 (1996)
Moral, S., Salmerón, A.: Dynamic importance sampling in Bayesian networks based on probability trees. Int. J. Approx. Reasoning 38(3), 245–261 (2005)
Dechter, R., Rish, I.: Mini-buckets: a general scheme for bounded inference. J. ACM (JACM) 50(2), 107–153 (2003)
Wainwright, M.J., Jordan, M.I., et al.: Graphical models, exponential families, and variational inference. Found. Trends Mach. Learn. 1(1–2), 1–305 (2008)
McEliece, R.J., MacKay, D.J.C., Cheng, J.F.: Turbo decoding as an instance of pearl’s “belief propagation” algorithm. IEEE J. Sel. Areas Commun. 16(2), 140–152 (1998)
Darwiche, A.: Modeling and Reasoning with Bayesian Networks. Cambridge University Press, Cambridge (2009)
Cano, A., Moral, S., Salmerón, A.: Novel strategies to approximate probability trees in penniless propagation. Int. J. Intell. Syst. 18(2), 193–203 (2003)
Kask, K., Dechter, R.: Branch and bound with mini-bucket heuristics. IJCAI 99, 426–433 (1999)
Kullback, S., Leibler, R.: On information and sufficiency. Ann. Math. Stat. 22, 76–86 (1951)
Kohlas, J., Wilson, N.: Semiring induced valuation algebras: exact and approximate local computation algorithms. Artif. Intell. 172(11), 1360–1399 (2008)
Acknowledgements
This research was supported by the Spanish Ministry of Economy and Competitiveness under project TIN2016-77902-C3-2-P and the European Regional Development Fund (FEDER).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG, part of Springer Nature
About this paper
Cite this paper
Moral, S. (2018). Divergence Measures and Approximate Algorithms for Valuation Based Systems. In: Medina, J., Ojeda-Aciego, M., Verdegay, J., Perfilieva, I., Bouchon-Meunier, B., Yager, R. (eds) Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications. IPMU 2018. Communications in Computer and Information Science, vol 855. Springer, Cham. https://doi.org/10.1007/978-3-319-91479-4_49
Download citation
DOI: https://doi.org/10.1007/978-3-319-91479-4_49
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-91478-7
Online ISBN: 978-3-319-91479-4
eBook Packages: Computer ScienceComputer Science (R0)