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

An accelerated learning algorithm for multilayer perceptron networks.

A G Parlos1, B Fernandez, A F Atiya

  • 1Dept. of Nucl. Eng., Texas AandM Univ., College Station, TX.

IEEE Transactions on Neural Networks
|January 1, 1994
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

Single-use, low-profile flexible ureterorenoscopy in paediatric lithiasis: Preliminary experience expanding the boundaries of ESPU recommendations.

Journal of pediatric surgery·2026
Same author

Extra-hepatic cholangiocarcinoma diagnosis: from classical pathological analysis to the emerging omics tests.

European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology·2026
Same author

Single-event fast neutron time-of-flight spectrometry with a petawatt-laser-driven neutron source.

Nature communications·2026
Same author

COSMONAUT: A COmpact spectrometer for measurements of neutrons at the ASDEX upgrade tokamak.

The Review of scientific instruments·2024
Same author

Floating plastics as integrative samplers of organic contaminants of legacy and emerging concern from Western Mediterranean coastal areas.

The Science of the total environment·2023
Same author

Deep ulcers are associated with increased C-reactive protein in active ulcerative colitis.

Digestive and liver disease : official journal of the Italian Society of Gastroenterology and the Italian Association for the Study of the Liver·2023
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

A new adaptive back propagation (ABP) algorithm speeds up supervised training for multilayer perceptron networks. This method enhances convergence for analog problems without extra tuning parameters.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Neural Networks

Background:

  • Supervised training of multilayer perceptron networks is crucial for many AI applications.
  • Traditional backpropagation algorithms can suffer from slow convergence and sensitivity to parameters.

Purpose of the Study:

  • To introduce an accelerated learning algorithm, adaptive back propagation (ABP), for supervised training of neural networks.
  • To improve convergence speed and reduce parameter sensitivity compared to existing methods.

Main Methods:

  • Developed an adaptive back propagation (ABP) algorithm inspired by "forced dynamics" for error functional.
  • Modified the learning rate as a specific function of error and error gradient norm.
  • Ensured no additional tuning parameters were introduced, unlike other backpropagation variants.

Related Experiment Videos

Main Results:

  • ABP demonstrated superior convergence speed for analog problems compared to competing methods.
  • The algorithm showed reduced sensitivity to variations in the step size parameter.
  • Effectiveness was validated through simulation results.

Conclusions:

  • The adaptive back propagation (ABP) algorithm offers a significant improvement in convergence speed for supervised neural network training, particularly for analog problems.
  • ABP provides a more robust and efficient training method by eliminating the need for extra tuning parameters and reducing sensitivity to step size.