Abstract
Given a collection of random variables [ξ i ] i ∈ N where N is a finite nonempty set, the corresponding multiinformation function assigns to each subset A ⊂ N the relative entropy of the joint distribution of [ξ i ] i ∈ A with respect to the product of distributions of individual random variables ξ i for i ∈ A. We argue that it is a useful tool for problems concerning stochastic (conditional) dependence and independence (at least in the discrete case).
First, the multiinformation function makes it possible to express the conditional mutual information between [ξ i ] i ∈ A and [ξ i ] i ∈ В given [ξ i ] i ∈ C (for every disjoint A,B, C ⊂ N), which can be considered as a good measure of conditional stochastic dependence. Second, one can introduce reasonable measures of dependence of level r among variables [ξ i ] i ∈ A (where A ⊂ N, 1 ≤ r < card A) which are expressible by means of the multiinformation function. Third, it enables one to derive theoretical results on (nonexistence of an) axiomatic characterization of stochastic conditional independence models.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
de Campos, L.M. (1995) Independence relationships in possibility theory and their application to learning in belief networks, in G. Della Riccia, R. Kruse and R. Viertl (eds.), Mathematical and Statistical Methods in Artificial Intelligence, Springer-Verlag, 119 – 130.
Csiszár, I. (1975) I-divergence geometry of probability distributions and minimazition problems, Ann. Probab., 3, 146 – 158.
Cover, T.M., and Thomas, J.A. (1991) Elements of Information Theory, John Wiley, New York.
Darroch, J.N., Lauritzen, S.L., and Speed, T.P. (1980) Markov fields and log-linear interaction models for contingency tables, Ann. Statist., 8, 522 – 539.
Dawid, A.P. (1979) Conditional independence in statistical theory, J. Roy. Stat. Soc. В, 41, 1 – 31.
Fonck P. (1994) Conditional independence in possibility theory, in R.L. de Mantaras and D. Poole (eds.), Uncertainty in Artificial Intelligence: proceedings of the 10th conference, Morgan Kaufman, San Francisco, 221 – 226.
Gallager, R.G. (1968) Information Theory and Reliable Communication, John Wiley, New York.
Han T.S. (1978) Nonnegative entropy of multivariate symmetric correlations, Information and Control, 36, 113 – 156.
Malvestuto, F.M. (1983) Theory of random observables in relational data bases, Inform. Systems, 8, 281 – 289.
Matúš, F., and Studený, M. (1995) Conditional independencies among four random variables I., Cоmbinatorics, Probability and Computing, 4, 269 – 278.
Matúš, F. (1995) Conditional independencies among four random variables II., Combinatorics, Probability and Computing, 4, 407 – 417.
Matúš, F. (1998) Conditional independencies among four random variables III., submitted to Combinatorics, Probability and Computing.
Pearl, J., and Paz, A. (1987) Graphoids: graph-based logic for reasoning about relevance relations, in B. Du Boulay, D. Hogg and L. Steels (eds.), Advances in Artificial Intelligence - II, North Holland, Amsterdam, pp. 357 – 363.
Pearl, J. (1988) Probabilistic Reasoning in Intelligent Systems: networks of plausible inference, Morgan Kaufmann, San Mateo.
Perez, A. (1977) ε-admissible simplifications of the dependence structure of a set of random variables, Kybernetika, 13, 439 – 449.
Rényi, A. (1959) On measures of dependence, Acta Math. Acad. Sci. Hung., 10, 441 – 451.
Spohn, W. (1980) Stochastic independence, causal independence and shieldability, J. Philos. Logic, 9, 73 – 99.
Studený, M. (1987) Asymptotic behaviour of empirical multiinformation, Kybernetika, 23, 124–135.
Studený, M. (1989) Multiinformation and the problem of characterization of conditional independence relations, Problems of Control and Information Theory, 18, 3 – 16.
Studený, M. (1992) Conditional independence relations have no finite complete characterization, in S. Kubík and J.Á. Víšek (eds.), Information Theory, Statistical Decision Functions and Random Processes: proceedings of the 11th Prague conference - B, Kluwer, Dordrecht (also Academia, Prague ), pp. 377 – 396.
Studený, M. (1987) The concept of multiinformation in probabilistic decision-making (in Czech), PhD. thesis, Institute of Information Theory and Automation, Czechoslovak Academy of Sciences, Prague.
Vejnarová, J. (1994) A few remarks on measures of uncertainty in Dempster-Shafer theory, Int. J. General Systems, 22, pp. 233 – 243.
Vejnarová J. (1997) Measures of uncertainty and independence concept in different calculi, accepted to EPIA’97.
Watanabe, S. (1960) Information theoretical analysis of multivariate correlation, IBM Journal of research and development, 4, pp. 66 – 81.
Watanabe, S. (1969) Knowing and Guessing: a qualitative study of inference and information, John Wiley, New York.
Xiang, Y., Wong, S.K.M., and Cercone, N. (1996) Critical remarks on single link search in learning belief networks, in E. Horvitz and F. Jensen (eds.), Uncertainty in Artificial Intelligence: proceedings of 12th conference, Morgan Kaufman, San Francisco, 564 – 571.
Zhang, Z., and Yeung, R. (1997) A non-Shannon type conditional information inequality, to appear in IEEE Transactions on Information Theory.
Zvárová, J. (1974) On measures of statistical dependence, Časopis pro pěstování matematiky, 99, 15 – 29.
Zvárová, J., and Studený, M. (1997) Information-theoretical approach to constitution and reduction of medical data, Int. J. Medical Informatics, 45, 65 – 74.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Studený, M., Vejnarová, J. (1998). The Multiinformation Function as a Tool for Measuring Stochastic Dependence. In: Jordan, M.I. (eds) Learning in Graphical Models. NATO ASI Series, vol 89. Springer, Dordrecht. https://doi.org/10.1007/978-94-011-5014-9_10
Download citation
DOI: https://doi.org/10.1007/978-94-011-5014-9_10
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-010-6104-9
Online ISBN: 978-94-011-5014-9
eBook Packages: Springer Book Archive