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

Mixed Markov models.

Arthur Fridman1

  • 1Applied Computer Science and Mathematics, Merck & Co., Inc., Rahway, NJ 07065, USA. arthur_fridman@merck.com

Proceedings of the National Academy of Sciences of the United States of America
|June 28, 2003
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

Phage Display-Based Discovery of HPV Capsomere-Specific Antibodies for HPV Process Characterization.

Biotechnology journal·2026
Same author

Characterization of humoral and cellular immunity induced by mRNA vaccines expressing norovirus VP1 proteins in mice and NHPs.

Molecular therapy : the journal of the American Society of Gene Therapy·2025
Same author

Stabilization of norovirus GII.3 virus-like particles by rational disulfide engineering.

NPJ vaccines·2025
Same author

Discovery of Antibodies Against Endemic Coronaviruses with NGS-Based Human Fab Phage Display Platform.

Antibodies (Basel, Switzerland)·2025
Same author

Comparison of Molnupiravir Exposure-Response Relationships for Virology Response and Mechanism of Action Biomarkers With Clinical Outcomes in Treatment of COVID-19.

Clinical and translational science·2025
Same author

Correction to: Virologic Outcomes with Molnupiravir in Nonhospitalized Adult Patients with COVID-19 from the Randomized, Placebo-Controlled MOVe-OUT Trial.

Infectious diseases and therapy·2024

Mixed Markov models offer a flexible alternative to traditional Markov random fields by allowing context-dependent interactions. This research explores their properties and efficient probabilistic inference methods.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Probabilistic Graphical Models

Background:

  • Markov random fields (MRFs) model complex probabilistic relationships but struggle with context-specific interactions due to rigid graph structures.
  • The requirement for fully connected graphs in MRFs can lead to computational intractability and obscure variable independencies.

Purpose of the Study:

  • To investigate the analytical and computational properties of Mixed Markov Models (MMMs).
  • To demonstrate the advantages of MMMs in representing context-dependent dependencies.
  • To develop efficient inference algorithms for MMMs.

Main Methods:

  • Introduced Mixed Markov Models (MMMs) with node-valued random variables that dynamically augment graph structures.
  • Analyzed the local and global Markov properties of positive mixed models.

Related Experiment Videos

  • Developed a computationally efficient procedure for probabilistic inference in MMMs.

    • Main Results:

    • Established that positive mixed models possess a local Markov property equivalent to their global factorization.
    • Demonstrated that MMMs can effectively encode context-specific interactions absent in traditional MRFs.
    • Presented a computationally efficient method for probabilistic query answering in MMMs.

    Conclusions:

    • Mixed Markov Models provide a powerful and flexible framework for probabilistic modeling, overcoming limitations of traditional MRFs.
    • The developed inference procedure enhances the practical applicability of MMMs in complex probabilistic systems.
    • Further research into MMMs can advance AI and machine learning applications requiring nuanced relational modeling.