Online estimation of hidden Markov models

We present a novel and simple online estimation algorithm for hidden Markov models, with memory requirements independent of the data length. The transition matrices and the state distribution are obtained at any instant as contractions of tensorial quantities, which are iteratively reestimated.

[1]  Ghassan Kawas Kaleh,et al.  Joint parameter estimation and symbol detection for linear or nonlinear unknown channels , 1994, IEEE Trans. Commun..

[2]  Geir Hovland,et al.  Hidden Markov Models as a Process Monitor in Robotic Assembly , 1998, Int. J. Robotics Res..

[3]  M. Borodovsky,et al.  GeneMark.hmm: new solutions for gene finding. , 1998, Nucleic acids research.

[4]  H. Furstenberg.  Noncommuting random products , 1963 .

[5]  Harold J. Kushner,et al.  Stochastic Approximation Algorithms and Applications , 1997, Applications of Mathematics.

[6]  H. Kushner.  Asymptotic global behavior for stochastic approximation and diffusions with slowly decreasing noise effects: Global minimization via Monte Carlo , 1987 .

[7]  John B. Moore,et al.  Hidden Markov Models: Estimation and Control , 1994 .

[8]  John B. Moore,et al.  Adaptive Estimation of Hmm Transition Probabilities , 1996, Fourth International Symposium on Signal Processing and Its Applications.

[9]  Giridharan Iyengar,et al.  Models for automatic classification of video sequences , 1997, Electronic Imaging.

[10]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[11]  Volker Breuer,et al.  Baum-Welch reestimation formula for coupled sequential machines , 1995, IEEE Trans. Signal Process..

[12]  T. Rydén.  On recursive estimation for hidden Markov models , 1997 .

[13]  Angelo Vulpiani,et al.  Products of Random Matrices , 1993 .

[14]  L. Baum,et al.  Statistical Inference for Probabilistic Functions of Finite State Markov Chains , 1966 .

[15]  S H Chung,et al.  Characterization of single channel currents using digital signal processing techniques based on Hidden Markov Models. , 1990, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[16]  Lalit R. Bahl,et al.  A Maximum Likelihood Approach to Continuous Speech Recognition , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.