Online expectation maximization based algorithms for inference in hidden Markov models
From MaRDI portal
Abstract: The Expectation Maximization (EM) algorithm is a versatile tool for model parameter estimation in latent data models. When processing large data sets or data stream however, EM becomes intractable since it requires the whole data set to be available at each iteration of the algorithm. In this contribution, a new generic online EM algorithm for model parameter inference in general Hidden Markov Model is proposed. This new algorithm updates the parameter estimate after a block of observations is processed (online). The convergence of this new algorithm is established, and the rate of convergence is studied showing the impact of the block size. An averaging procedure is also proposed to improve the rate of convergence. Finally, practical illustrations are presented to highlight the performance of these algorithms in comparison to other online maximum likelihood procedures.
Recommendations
- Convergence of a particle-based approximation of the block online expectation maximization algorithm
- On-line expectation-maximization algorithm for latent data models
- Online Learning with Hidden Markov Models
- Efficient inference in state-space models through adaptive learning in online Monte Carlo expectation maximization
- On-line learning for the infinite hidden Markov model
Cites work
- Acceleration of Stochastic Approximation by Averaging
- Analyticity, Convergence, and Convergence Rate of Recursive Maximum-Likelihood Estimation in Hidden Markov Models
- Asymptotic properties of the maximum likelihood estimator in autoregressive models with Markov regime
- Convergence of a particle-based approximation of the block online expectation maximization algorithm
- Convergence of the Monte Carlo expectation maximization for curved exponential families.
- Hidden Markov chains and the analysis of genome structure
- Hidden Markov Models for Speech Recognition
- scientific article; zbMATH DE number 3723610 (Why is no real title available?)
- scientific article; zbMATH DE number 3567782 (Why is no real title available?)
- scientific article; zbMATH DE number 1043533 (Why is no real title available?)
- Large deviations for interacting particle systems: Applications to non-linear filtering
- Markov chains and stochastic stability
- Non-asymptotic deviation inequalities for smoothed additive functionals in nonlinear state-space models
- On contraction properties of Markov kernels
- On-line expectation-maximization algorithm for latent data models
- Online expectation maximization based algorithms for inference in hidden Markov models
- Online Learning with Hidden Markov Models
- Sequential Monte Carlo Methods in Practice
- Stochastic Limit Theory
Cited in
(25)- Online EM for functional data
- Online expectation maximization based algorithms for inference in hidden Markov models
- Asymptotic analysis of model selection criteria for general hidden Markov models
- Online inference with multi-modal likelihood functions
- Online learning of Riemannian hidden Markov models in homogeneous Hadamard spaces
- Efficient inference in state-space models through adaptive learning in online Monte Carlo expectation maximization
- Recursive estimation of multivariate hidden Markov model parameters
- On particle methods for parameter estimation in state-space models
- On-line expectation-maximization algorithm for latent data models
- On-line learning for the infinite hidden Markov model
- Online Learning with Hidden Markov Models
- Convergence of a particle-based approximation of the block online expectation maximization algorithm
- On-Line Inference for Hidden Markov Models via Particle Filters
- Online Learning of Inverted Beta-Liouville HMMs for Anomaly Detection in Crowd Scenes
- On the two-filter approximations of marginal smoothing distributions in general state-space models
- Statistical inference for oscillation processes
- Offline and online identification of hidden semi-Markov models
- Approximate, computationally efficient online learning in Bayesian spiking neurons
- Global implicit function theorems and the online expectation–maximisation algorithm
- Divide-and-conquer Bayesian inference in hidden Markov models
- Asynchronous and Distributed Data Augmentation for Massive Data Settings
- Identifiability of discrete input–output hidden Markov models with external signals
- A new algorithm for inference in HMM's with lower span complexity
- Distributed online expectation-maximization algorithm for Poisson mixture model
- Accelerated distributed expectation-maximization algorithms for the parameter estimation in multivariate Gaussian mixture models
This page was built for publication: Online expectation maximization based algorithms for inference in hidden Markov models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1951134)
