Regenerativity of Viterbi process for pairwise Markov models

DOI10.1007/S10959-020-01022-ZzbMATH Open1459.62158arXiv2103.11821OpenAlexW3137362470MaRDI QIDQ2224946FDOQ2224946

Authors: J. Lember, Joonas Sova 

Publication date: 4 February 2021

Published in: Journal of Theoretical Probability (Search for Journal in Brave)

Abstract: For hidden Markov models one of the most popular estimates of the hidden chain is the Viterbi path -- the path maximising the posterior probability. We consider a more general setting, called the pairwise Markov model (PMM), where the joint process consisting of finite-state hidden process and observation process is assumed to be a Markov chain. It has been recently proven that under some conditions the Viterbi path of the PMM can almost surely be extended to infinity, thereby defining the infinite Viterbi decoding of the observation sequence, called the Viterbi process. This was done by constructing a block of observations, called a barrier, which ensures that the Viterbi path goes trough a given state whenever this block occurs in the observation sequence. In this paper we prove that the joint process consisting of Viterbi process and PMM is regenerative. The proof involves a delicate construction of regeneration times which coincide with the occurrences of barriers. As one possible application of our theory, some results on the asymptotics of the Viterbi training algorithm are derived.

Full work available at URL: https://arxiv.org/abs/2103.11821

Recommendations

• Existence of infinite Viterbi path for pairwise Markov models
• scientific article; zbMATH DE number 7424978
• On the Viterbi process with continuous state space
• Infinite Viterbi alignments in the two state hidden Markov models
• Properties of the maximum a posteriori path estimator in hidden Markov models

zbMATH Keywords

Viterbi algorithmhidden Markov modelMarkov switching modelViterbi trainingMAP pathViterbi path

Mathematics Subject Classification ID

Markov processes: estimation; hidden Markov models (62M05) Markov processes: hypothesis testing (62M02) Signal detection and filtering (aspects of stochastic processes) (60G35) Foundations of stochastic processes (60G05)

Cites Work

• Model-Based Clustering, Discriminant Analysis, and Density Estimation
• A classification EM algorithm for clustering and two stochastic versions
• Markov Chains and Stochastic Stability
• A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle
• Hidden Markov processes
• Inference in hidden Markov models.
• Real Analysis and Probability
• Statistical Inference for Probabilistic Functions of Finite State Markov Chains
• Title not available (Why is that?)
• Analysis of time series subject to changes in regime
• A decision-theoretic approach for segmental classification
• The adjusted Viterbi training for hidden Markov models
• Bridging Viterbi and posterior decoding: a generalized risk approach to hidden path inference based on hidden Markov models
• On the accuracy of the MAP inference in HMMs
• Infinite Viterbi alignments in the two state hidden Markov models
• Asymptotic risks of Viterbi segmentation
• A Constructive Proof of the Existence of Viterbi Processes
• On approximation of smoothing probabilities for hidden Markov models
• Title not available (Why is that?)
• Properties of the maximum a posteriori path estimator in hidden Markov models
• On the Viterbi process with continuous state space
• Convergence of the maximum a posteriori path estimator in hidden Markov models
• Large Deviation Bounds for Functionals of Viterbi Paths
• ADJUSTED VITERBI TRAINING
• On adjusted Viterbi training
• Signal and Image Segmentation Using Pairwise Markov Chains
• The segmental K-means algorithm for estimating parameters of hidden Markov models
• Unsupervised data classification using pairwise Markov chains with automatic copulas selection
• Existence of infinite Viterbi path for pairwise Markov models

Cited In (8)

• Viterbi sequences and polytopes
• Pairwise Markov models and hybrid segmentation approach
• ADJUSTED VITERBI TRAINING
• The infinite viterbi alignment and decay-convexity
• Exponential forgetting of smoothing distributions for pairwise Markov models
• Viterbi algorithms for hidden Markov models with partially visible states
• Existence of infinite Viterbi path for pairwise Markov models
• A Constructive Proof of the Existence of Viterbi Processes

This page was built for publication: Regenerativity of Viterbi process for pairwise Markov models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2224946)