Efficient and effective learning of HMMs based on identification of hidden states

DOI10.1155/2017/7318940MaRDI QIDQ1993090zbMATH OpenOpenAlexWikidataFDO

Authors Tingting Liu, Jan Lemeire

Publication date 5 November 2018

Published in Mathematical Problems in Engineering (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://doi.org/10.1155/2017/7318940

Mathematics Subject Classification ID

Markov processes: estimation; hidden Markov models (62M05) Identification in stochastic control theory (93E12)

Review text

Summary: The predominant learning algorithm for Hidden Markov Models (HMMs) is local search heuristics, of which the Baum-Welch (BW) algorithm is mostly used. It is an iterative learning procedure starting with a predefined size of state spaces and randomly chosen initial parameters. However, wrongly chosen initial parameters may cause the risk of falling into a local optimum and a low convergence speed. To overcome these drawbacks, we propose to use a more suitable model initialization approach, a Segmentation-Clustering and Transient analysis (SCT) framework, to estimate the number of states and model parameters directly from the input data. Based on an analysis of the information flow through HMMs, we demystify the structure of models and show that high-impact states are directly identifiable from the properties of observation sequences. States having a high impact on the log-likelihood make HMMs highly specific. Experimental results show that even though the identification accuracy drops to 87.9\% when random models are considered, the SCT method is around 50 to 260 times faster than the BW algorithm with 100\% correct identification for highly specific models whose specificity is greater than 0.06.

Recommendations

• A spectral algorithm for learning hidden Markov models
• Discriminative learning between the states of the hidden Markov model for speech recognition
• A sequential pruning strategy for the selection of the number of states in hidden Markov models
• A low complexity simulated annealing approach for training hidden Markov models
• Tensor train spectral method for learning of hidden Markov models (HMM)

Cites work

• A Maximization Technique Occurring in the Statistical Analysis of Probabilistic Functions of Markov Chains
• A spectral algorithm for learning hidden Markov models
• Detecting Homogeneous Segments in DNA Sequences by Using Hidden Markov Models
• Equivalence of state representations for hidden Markov models
• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 2019610 (Why is no real title available?)
• scientific article; zbMATH DE number 3444596 (Why is no real title available?)
• scientific article; zbMATH DE number 3340881 (Why is no real title available?)
• Statistical Inference for Probabilistic Functions of Finite State Markov Chains
• Sticky Hidden Markov Modeling of Comparative Genomic Hybridization

Cited in

(3)

• Learning hidden Markov models with unknown number of states
• Temporal BYY learning for state space approach, hidden Markov model, and blind source separation
• Initialization of Hidden Markov and Semi‐Markov Models: A Critical Evaluation of Several Strategies

This page was built for publication: Efficient and effective learning of HMMs based on identification of hidden states

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1993090)