Jove
Visualize
Contact Us

Related Concept Videos

Newton’s Method01:30

Newton’s Method

Newton’s Method is a powerful iterative technique for approximating the roots of real-valued, differentiable functions, particularly when analytical solutions are impractical. This approach is widely used in scientific computing, engineering, and finance, where equations may be too complex for traditional algebraic methods to handle. The method relies on an iterative process that refines an initial estimate using the function’s derivative to approach the true solution progressively.
Alternating Series and Absolute Convergence01:28

Alternating Series and Absolute Convergence

A mass attached to a vertical spring can exhibit oscillatory motion as it moves above and below a central equilibrium point. In an ideal spring, the oscillations would continue indefinitely with constant amplitude. In a damped spring, however, resistive forces such as air resistance or internal friction gradually reduce the size of each swing. This behavior is often modeled by combining a sinusoidal function, which represents the repeated motion, with an exponential decay factor, which reduces...
Convergence of Sequences01:26

Convergence of Sequences

A sequence is a function defined on the natural numbers that assigns a value to each index. It can be understood as an ordered list of terms generated one after another. In mathematical analysis, an important question is whether the terms of a sequence approach a single real number as the index becomes very large. When this happens, the sequence is said to converge, and the value approached is called the limit. From a graphical perspective, convergence means that the plotted terms approach a...
Geometric Sequences01:30

Geometric Sequences

In systems where values diminish by a constant proportion at each stage, the resulting sequence follows a geometric structure. Each new value in the sequence is obtained by applying a fixed multiplier to the preceding term. This regular, proportional decline type is often used to represent processes involving gradual loss, such as energy dissipation or reduction in amplitude over time.When analyzing the total effect of such a process across unlimited iterations, the series of values is referred...
Convergence of Taylor Series01:30

Convergence of Taylor Series

The Taylor series provides a systematic method for approximating a smooth function by a polynomial that closely matches the function near a chosen point. This approach is particularly valuable in scientific and engineering contexts where functions may be difficult to evaluate directly, such as oscillatory voltages in alternating current (AC) circuits. Replacing complex functions with polynomial expressions simplifies computation while preserving essential local behavior. Taylor’s Theorem...
Newton's Third Law: Introduction00:58

Newton's Third Law: Introduction

Whenever one body exerts a force on a second body, the first body experiences a force equal in magnitude and opposite in direction, to the force that it exerts. For instance, when a person pushes on a wall, the wall exerts an equal and opposite force towards the person. This brings us to Newton's third law of motion. Newton's third law represents a certain symmetry in nature: Forces always occur in pairs, and one body cannot exert a force on another without experiencing a force itself. This law...

You might also read

Related Articles

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

Sort by
Same author

Convergence of learning algorithms with constant learning rates.

IEEE transactions on neural networks·1991
See all related articles
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 Video

Updated: Jul 7, 2026

Design and Application of a Fault Detection Method Based on Adaptive Filters and Rotational Speed Estimation for an Electro-Hydrostatic Actuator
06:45

Design and Application of a Fault Detection Method Based on Adaptive Filters and Rotational Speed Estimation for an Electro-Hydrostatic Actuator

Published on: October 28, 2022

A recurrent Newton algorithm and its convergence properties.

C M Kuan1

  • 1Dept. of Econ., Nat. Taiwan Univ., Taipei.

IEEE Transactions on Neural Networks
|January 1, 1995
PubMed
Summary
This summary is machine-generated.

A new recurrent Newton algorithm for recurrent neural networks was developed. This algorithm, with a specific constraint, shows superior performance over backpropagation and unconstrained methods in reducing mean-squared errors.

Related Experiment Videos

Last Updated: Jul 7, 2026

Design and Application of a Fault Detection Method Based on Adaptive Filters and Rotational Speed Estimation for an Electro-Hydrostatic Actuator
06:45

Design and Application of a Fault Detection Method Based on Adaptive Filters and Rotational Speed Estimation for an Electro-Hydrostatic Actuator

Published on: October 28, 2022

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Recurrent neural networks (RNNs) are crucial for sequential data processing.
  • Training RNNs effectively, particularly with complex dynamics, remains a challenge.
  • Existing training algorithms like backpropagation may suffer from convergence issues.

Purpose of the Study:

  • Introduce a novel recurrent Newton algorithm for a specific class of RNNs.
  • Address convergence challenges in training RNNs.
  • Improve the performance of RNN training algorithms.

Main Methods:

  • Developed a recurrent Newton algorithm incorporating a constraint on recurrent variables.
  • Compared the proposed algorithm against the standard backpropagation algorithm.
  • Evaluated performance using mean-squared errors in simulations.

Main Results:

  • The proposed recurrent Newton algorithm with the constraint demonstrates superior performance.
  • Achieved uniformly better results compared to the unconstrained Newton algorithm.
  • Outperformed the backpropagation algorithm in terms of mean-squared errors.

Conclusions:

  • Imposing a suitable constraint is essential for the convergence of the recurrent Newton algorithm.
  • The constrained recurrent Newton algorithm offers a more effective training method for RNNs.
  • This advancement has implications for improving deep learning models that utilize RNNs.