Exponential bounds for convergence of entropy rate approximations in hidden Markov models satisfying a path-mergeability condition

DOI10.1016/J.SPA.2014.07.011MaRDI QIDQ744238zbMATH OpenOpenAlexFDO

Authors Nicholas F. Travers

Publication date 6 October 2014

Published in Stochastic Processes and their Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1211.6181

zbMATH Keywords

hidden Markov modelexponential convergenceentropy ratepath-mergeable states

Mathematics Subject Classification ID

Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Measures of information, entropy (94A17)

Abstract: A hidden Markov model (HMM) is said to have path-mergeable states if for any two states i,j there exists a word w and state k such that it is possible to transition from both i and j to k while emitting w. We show that for a finite HMM with path-mergeable states the block estimates of the entropy rate converge exponentially fast. We also show that the path-mergeability property is asymptotically typical in the space of HMM topolgies and easily testable.

Recommendations

• On the entropy of a hidden Markov process
• Convergence of the maximum a posteriori path estimator in hidden Markov models
• Analyticity of Entropy Rate of Hidden Markov Chains
• Bounds on the entropy rate of binary hidden Markov processes
• Entropy rate for hidden Markov chains with rare transitions

Cites work

• scientific article; zbMATH DE number 3139241 (Why is no real title available?)
• scientific article; zbMATH DE number 1517989 (Why is no real title available?)
• Analyticity of Entropy Rate of Hidden Markov Chains
• Approximations for the Entropy for Functions of Markov Chains
• Asymptotic synchronization for finite-state sources
• Discrete-time filtering for linear systems with non-Gaussian initial conditions: asymptotic behavior of the difference between the MMSE and LMSE estimates
• Elements of Information Theory
• Exact synchronization for finite-state sources
• Exponential forgetting and geometric ergodicity in hidden Markov models
• Exponential stability for nonlinear filtering
• Forgetting the initial distribution for hidden Markov models
• Hidden Markov Models for Speech Recognition
• Loss of memory of hidden Markov models and Lyapunov exponents
• Lyapunov Exponents for Finite State Nonlinear Filtering
• On the Identifiability Problem for Functions of Finite Markov Chains
• On the Identifiability Problem for Functions of Finite Markov Chains
• Regularities unseen, randomness observed: Levels of entropy convergence
• Stability of nonlinear filters in nonmixing case
• State learning and mixing in entropy of hidden Markov processes and the Gilbert-Elliott channel

Cited in

(2)

• Shannon entropy rate of hidden Markov processes
• Spectral simplicity of apparent complexity. I: The nondiagonalizable metadynamics of prediction

This page was built for publication: Exponential bounds for convergence of entropy rate approximations in hidden Markov models satisfying a path-mergeability condition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q744238)