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

CMAC with General Basis Functions.

Lin Chun-Shin1, Chiang Ching-Tsan

  • 1University of Missouri-Columbia, Missouri, USA

Neural Networks : the Official Journal of the International Neural Network Society
|October 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 journal

Unified knowledge distillation: Integrating offline, online, and self-distillation paradigms for image classification.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

A survey of modern neural network based methods for the graph coloring problem.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Convergence of stochastic gradient methods for wide two-layer physics-informed neural networks for the Poisson equation.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

CANav: Cognition-aligned object-goal navigation based on a hierarchical scene graph with personalized knowledge-guided reasoning.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Predefined-time synchronization of memristor-based competitive neural networks with time-varying delays and application in image encryption.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

UMind: A unified multitask network for zero-shot M/EEG visual decoding.

Neural networks : the official journal of the International Neural Network Society·2026

This study enhances the cerebellar model articulation controller (CMAC) by replacing constant basis functions with differentiable ones. This modification improves modeling accuracy and preserves derivative information for advanced learning control applications.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Control Systems Engineering

Background:

  • The cerebellar model articulation controller (CMAC) is a widely used learning control technique, often conceptualized as a basis function network (BFN).
  • Conventional CMAC employs local constant basis functions, leading to limitations such as constant output within quantized states and loss of derivative information.

Purpose of the Study:

  • To investigate a generalized CMAC scheme using non-constant, differentiable basis functions.
  • To address the limitations of conventional CMAC regarding output continuity and derivative information preservation.

Main Methods:

  • Developed a generalized CMAC framework incorporating general differentiable basis functions.
  • Derived the mathematical foundation for the modified CMAC scheme.

Related Experiment Videos

  • Proved the convergence of the learning process for the generalized CMAC.
  • Conducted simulations using Gaussian basis functions (GBFs) within the generalized CMAC framework.

    • Main Results:

    • Demonstrated that the generalized CMAC scheme encompasses the conventional CMAC as a special case.
    • Established the mathematical basis and proved learning convergence for the modified CMAC.
    • Simulations with Gaussian basis functions showed enhanced modeling accuracy compared to conventional CMAC.
    • Verified the capability of the modified CMAC to store and utilize derivative information.

    Conclusions:

    • The generalized CMAC with differentiable basis functions offers significant improvements in modeling accuracy and derivative information preservation.
    • This enhanced CMAC provides a more robust and capable learning control solution compared to traditional methods.
    • The findings pave the way for more sophisticated applications of CMAC in complex control tasks.