Abstract
Hidden Markov Processes (HMP) is one of the basic tools of the modern probabilistic modeling. The characterization of their entropy remains however an open problem. Here the entropy of HMP is calculated via the cycle expansion of the zeta-function, a method adopted from the theory of dynamical systems. For a class of HMP this method produces exact results both for the entropy and the moment-generating function. The latter allows to estimate, via the Chernoff bound, the probabilities of large deviations for the HMP. More generally, the method offers a representation of the moment-generating function and of the entropy via convergent series.
Similar content being viewed by others
References
Rabiner, L.R.: Proc. IEEE 77, 257–286 (1989)
Ephraim, Y., Merhav, N.: IEEE Trans. Inf. Theory 48, 1518–1569 (2002)
Crouse, M., Nowak, R., Baraniuk, R.: IEEE Trans. Signal Process. 46, 886 (1998)
Koski, T.: Hidden Markov Models for Bioinformatics. Kluwer Academic Publishers, Dordrecht (2001)
Baldi, P., Brunak, S.: Bioinformatics. MIT Press, Cambridge (2001)
Ash, R.: Information Theory. Interscience Publishers, New York (1965)
Cover, T.M., Thomas, J.A.: Elements of Information Theory. Wiley, New York (1991)
Blackwell, D.: The entropy of functions of finite-state Markov chains. In: Trans. First Prague Conf. Inf. Theory, Statistical Decision Functions, Random Processes, p. 13. Pub. House Chechoslovak Acad. Sci., Prague (1957)
Stratonovich, R.L.: Information Theory. Sovietskoe Radio, Moscow (1976). (in Russian)
Rezaeian, M.: Hidden Markov process: a new representation, entropy rate and estimation entropy. arXiv:cs.IT/0606114 (2006)
Birch, I.J.: Ann. Math. Stat. 33, 930 (1962)
Jacquet, P., Seroussi, G., Szpankowski, W.: On the entropy of a hidden Markov process. In: Int. Symp. Inf. Theory, p. 10. Chicago, IL, 2004
Holliday, T., Goldsmith, A., Glynn, P.: IEEE Trans. Inf. Theory 52, 3509 (2006)
Ordentlich, E., Weissman, T.: IEEE Trans. Inf. Theory 52, 19 (2006)
Egner, S. et al.: On the entropy rate of a hidden Markov model. In: Int. Symp. Inf. Theory, p. 12. Chicago, IL, 2004
Zuk, O., Kanter, I., Domany, E.: J. Stat. Phys. 121, 343 (2005)
Zuk, O., Kanter, I., Domany, E., Aizenman, M.: IEEE Signal Process. Lett. 13, 517 (2006)
Chigansky, P.: The entropy rate of a binary channel with slowly varying input. arXiv:cs/0602074
Horn, R.A., Johnson, C.R.: Matrix Analysis. Cambridge University Press, New Jersey (1985)
Kingman, J.F.C.: Ann. Probab. 1, 883 (1973)
Steele, J.M.: Ann. de l’I.H.P. B 25, 93 (1989)
Crisanti, A., Paladin, G., Vulpiani, A.: Products of random matrices in statistical physics. In: Springer Series in Solid State Sciences, vol. 104. Springer, Berlin (1993)
Goldsheid, L.Y., Margulis, G.A.: Russ. Math. Surv. 44, 11 (1989)
Orszag, S.A., Sulem, P.L., Goldirsch, I.: Physica D 27, 311 (1987)
Kontorovich, L.: Measure concentration of hidden Markov processes. arXiv:math/0608064 (2006)
Artuso, R., Aurell, E., Cvitanovic, P.: Nonlinearity 3, 325 (1990)
Cvitanovic, P.: Phys. Rev. Lett. 61, 2729 (1988)
Ruelle, D.: Statistical Mechanics, Thermodynamic Formalism. Addison-Wesley, Reading (1978)
Mainieri, R.: Chaos 2, 91 (1992)
Aurell, E.: J. Stat. Phys. 58, 967 (1990)
Nielsen, J.: Lyapunov exponents for products of random matrices. Available at http://citeseer.ist.psu.edu/438423.html
Arnold, L., Gundlach, V.M., Demetrius, L.: Ann. Appl. Probab. 4, 859 (1994)
Peres, Y.: Ann. Inst. H. Poincare Probab. Statist. 28, 131 (1992)
Han, G., Markus, B.: IEEE Trans. Inf. Theory 52, 5251 (2006)
Gurvits, L., Ledoux, J.: Linear Algebra and Applications 404, 85 (2005)
Petersen, K.: Lectures on Ergodic Theory. Available from http://www.math.unc.edu/Faculty/petersen/lecturespdf.pdf
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Allahverdyan, A.E. Entropy of Hidden Markov Processes via Cycle Expansion. J Stat Phys 133, 535–564 (2008). https://doi.org/10.1007/s10955-008-9613-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10955-008-9613-0