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

Local linear perceptrons for classification.

E Alpaydin1, M I Jordan

  • 1Dept. of Comput. Eng., Bogazici Univ., Istanbul.

IEEE Transactions on Neural Networks
|January 1, 1996
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

Machine learning: Trends, perspectives, and prospects.

Science (New York, N.Y.)·2015
Same author

Computational Consequences of a Bias toward Short Connections.

Journal of cognitive neuroscience·2013
Same author

A model of the learning of arm trajectories from spatial deviations.

Journal of cognitive neuroscience·2013
Same author

Omnivariate decision trees.

IEEE transactions on neural networks·2008
Same author

Constructive feedforward ART clustering networks. I.

IEEE transactions on neural networks·2008
Same author

Constructive feedforward ART clustering networks. II.

IEEE transactions on neural networks·2008
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces local linear perceptrons for pattern recognition, showing coupled learning improves performance. These models offer better generalization and efficiency than radial basis functions and multilayer perceptrons.

Area of Science:

  • Machine Learning
  • Pattern Recognition
  • Artificial Neural Networks

Background:

  • Investigates a novel structure using local linear perceptrons to approximate global class discriminants.
  • Explores cooperative and competitive combination strategies for these local linear models.

Purpose of the Study:

  • To evaluate the efficacy of coupled supervised learning for local model positions and linear mappings.
  • To compare cooperative and competitive local linear models against multilayer perceptrons and radial basis functions (RBFs).

Main Methods:

  • Developed cooperative and competitive models with coupled supervised learning for perceptron positions and mappings.
  • Utilized cross-entropy for goodness criteria and derived learning equations.
  • Compared performance on handwritten digit recognition using generalization accuracy, learning time, and parameter count.

Related Experiment Videos

Main Results:

  • Coupled supervised learning for local model parameters is superior to uncoupled approaches.
  • Local linear models demonstrate significantly better generalization and reduced memory usage compared to RBFs.
  • Local models exhibit faster learning and comparable or superior generalization to multilayer perceptrons with similar parameter counts.

Conclusions:

  • Local linear perceptron structures are promising for high-dimensional pattern recognition tasks.
  • These models offer a compelling alternative to existing methods like RBFs and MLPs, balancing accuracy, speed, and efficiency.