Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Online learning with hidden markov models.

Gianluigi Mongillo1, Sophie Deneve

  • 1Group for Neural Theory, Department d'Etudes Cognitives, Ecole Normale Supérieure, Collège de France, Paris 75006, France. Gianluigi.Mongillo@ens.fr

Neural Computation
|February 8, 2008
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Conspiracy beliefs and perceptual inference in times of political uncertainty.

Scientific reports·2024
Same author

Functional connectivity and glutamate levels of the medial prefrontal cortex in schizotypy are related to sensory amplification in a probabilistic reasoning task.

NeuroImage·2023
Same author

Short-Term Synaptic Plasticity: Microscopic Modelling and (Some) Computational Implications.

Advances in experimental medicine and biology·2022
Same author

From hallucinations to synaesthesia: A circular inference account of unimodal and multimodal erroneous percepts in clinical and drug-induced psychosis.

Neuroscience and biobehavioral reviews·2022
Same author

No increased circular inference in adults with high levels of autistic traits or autism.

PLoS computational biology·2021
Same author

Efficient and robust coding in heterogeneous recurrent networks.

PLoS computational biology·2021
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

We developed an online expectation-maximization (EM) algorithm for hidden Markov models (HMMs). This method efficiently estimates parameters in dynamic environments, offering an alternative to batch processing.

Area of Science:

  • Computational Neuroscience
  • Machine Learning

Background:

  • Hidden Markov Models (HMMs) are widely used for sequential data analysis.
  • Parameter estimation in HMMs typically relies on batch Expectation-Maximization (EM) algorithms.
  • Batch methods can be computationally intensive and unsuitable for real-time applications.

Purpose of the Study:

  • To introduce an online version of the EM algorithm for HMMs.
  • To enable recursive computation of sufficient statistics for parameter estimation.
  • To adapt HMM parameter estimation for dynamic environments and changing data statistics.

Main Methods:

  • Developed a recursive computational scheme for sufficient statistics, avoiding the batch forward-backward procedure.
  • Generalized the online EM algorithm to handle time-varying model parameters using a discount factor.

Related Experiment Videos

  • Demonstrated equivalence to the batch EM algorithm under specific conditions.

    • Main Results:

    • The online EM algorithm computes parameters recursively, suitable for streaming data.
    • The algorithm effectively handles environments where data statistics change over time.
    • Introduced a discount factor to manage parameter changes and algorithm convergence.

    Conclusions:

    • The online EM algorithm provides an efficient and adaptive approach for HMM parameter estimation.
    • This method is particularly valuable for probabilistic modeling in dynamic systems, such as neuroscience.
    • The online approach offers a viable alternative to traditional batch methods for real-time analysis.