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

Learning non-stationary conditional probability distributions.

D Husmeier1

  • 1Biomathematics and Statistics Scotland at the Scottish Crop Research Institute, Invergowrie, Dundee, UK. dirk@bioss.sari.ac.uk

Neural Networks : the Official Journal of the International Neural Network Society
|August 11, 2000
PubMed
Summary

Sophisticated neural networks and graphical models can predict probabilities in changing environments. However, enhanced training schemes are crucial for their practical application in non-stationary settings.

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

Optimal estimation of drift and diffusion coefficients in the presence of static localization error.

Physical review. E·2019
Same author

Reverse engineering of genetic networks with Bayesian networks.

Biochemical Society transactions·2003
Same author

An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers.

Neural networks : the official journal of the International Neural Network Society·2003
Same author

Detection of recombination in DNA multiple alignments with hidden Markov models.

Journal of computational biology : a journal of computational molecular cell biology·2001
Same author

Probabilistic divergence measures for detecting interspecies recombination.

Bioinformatics (Oxford, England)·2001
Same author

The Bayesian evidence scheme for regularizing probability-density estimating neural networks.

Neural computation·2000

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Statistical Modeling

Background:

  • Non-stationary environments pose challenges for predictive models.
  • Existing neural networks and graphical models show promise but require refinement.
  • Practical viability of advanced models is hindered by training limitations.

Purpose of the Study:

  • To highlight the need for improved training schemes in predictive modeling.
  • To address the limitations of current approaches in non-stationary environments.
  • To pave the way for more practical applications of sophisticated models.

Main Methods:

  • Review of current neural network architectures.
  • Analysis of graphical model capabilities.
  • Evaluation of existing training methodologies.

Related Experiment Videos

Main Results:

  • Sophisticated models demonstrate potential for conditional probability prediction.
  • Current training schemes are insufficient for non-stationary environments.
  • Significant improvements in training are necessary for practical viability.

Conclusions:

  • Further research into advanced training schemes is essential.
  • Bridging the gap between theoretical potential and practical application is key.
  • Enhanced training methods will unlock the full capabilities of these models.