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

A fast multilayer neural-network training algorithm based on the layer-by-layer optimizing procedures.

G J Wang1, C C Chen

  • 1Dept. of Mech. Eng., Nat. Chung-Hsing Univ., Taichung.

IEEE Transactions on Neural Networks
|January 1, 1996
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Methods of Medium Optimization01:28

Methods of Medium Optimization

Optimizing growth media enhances microbial proliferation and maximizes product yield. Statistical experimental design methodologies provide structured and reproducible approaches, offering progressively higher levels of robustness and efficiency.The One-Factor-at-a-Time (OFAT) MethodThe One-Factor-at-a-Time (OFAT) method involves adjusting a single variable while keeping all others constant. However, it cannot detect interactions between variables, often leading to suboptimal outcomes when...
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving01:29

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Morphine augmentation increases gallbladder visualization in patients pretreated with cholecystokinin.

Journal of nuclear medicine : official publication, Society of Nuclear Medicine·1997
Same author

Right-sided cardiac tumors detected by transesophageal echocardiography and its usefulness in differentiating the benign from the malignant ones.

The American journal of cardiology·1997
Same author

Postoperative MRI of anorectal malformation.

Journal of the Formosan Medical Association = Taiwan yi zhi·1997
Same author

Value of somatostatin receptor scintigraphy: a prospective study in gastrinoma of its effect on clinical management.

Gastroenterology·1997
Same author

Blood levels of reduced haloperidol versus clinical efficacy and extrapyramidal side effects of haloperidol.

Progress in neuro-psychopharmacology & biological psychiatry·1997
Same author

The binge eating and emotional change in a patient with postinfectious encephalomyelitis.

The Kaohsiung journal of medical sciences·1997
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 novel learning algorithm enhances multilayer feedforward neural network training by optimizing weight matrices and layer outputs. This new method significantly improves convergence speed and reduces computation time for dynamic system identification.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Neural Networks

Background:

  • Multilayer feedforward neural networks (MFNNs) are widely used in various machine learning tasks.
  • Training MFNNs involves adjusting network weights to minimize errors, which can be computationally intensive.
  • Existing learning algorithms may face challenges in terms of speed and efficiency, especially in dynamic environments.

Purpose of the Study:

  • To introduce a novel, faster learning algorithm for adjusting the weights of multilayer feedforward neural networks.
  • To enhance the efficiency and performance of neural network training processes.
  • To improve the capabilities of neural networks in dynamic system identification.

Main Methods:

  • The proposed algorithm treats the output layer's weight matrix (W(2)) and the previous layer's output vector (Y) as variable sets.

Related Experiment Videos

  • It identifies an optimal solution pair (W(2)*, Y(P)*) to minimize the sum-square-error for input patterns.
  • The same optimization method is applied to hidden layers, incorporating dynamic forgetting factors for enhanced system identification.

    • Main Results:

    • Computer simulations demonstrate that the new algorithm significantly outperforms existing learning algorithms.
    • The proposed method shows superior converging speed compared to traditional approaches.
    • The algorithm requires less computation time, indicating increased efficiency.

    Conclusions:

    • The novel learning algorithm offers a substantial improvement in training speed and computational efficiency for MFNNs.
    • Its effectiveness is particularly pronounced in dynamic system identification tasks due to the integration of dynamic forgetting factors.
    • This research presents a powerful new tool for accelerating and optimizing neural network applications.