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 Concept Videos

Linear Approximation in Frequency Domain01:26

Linear Approximation in Frequency Domain

Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear.
Linearization and Approximation01:26

Linearization and Approximation

Linearization is a mathematical technique used to approximate complex, nonlinear functions with simpler linear models in the vicinity of a chosen reference point. The method is based on the idea that, although a function may be difficult to evaluate exactly, its behavior near a specific input value can often be closely approximated by the tangent line at that point. This approach is particularly useful when small deviations from a known value are involved.Consider the square root function, for...
Linear Approximation in Time Domain01:21

Linear Approximation in Time Domain

Nonlinear systems often require sophisticated approaches for accurate modeling and analysis, with state-space representation being particularly effective. This method is especially useful for systems where variables and parameters vary with time or operating conditions, such as in a simple pendulum or a translational mechanical system with nonlinear springs.
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length, the...
Feedback control systems01:26

Feedback control systems

Feedback control systems are categorized in various ways based on their design, analysis, and signal types.
Linear feedback systems are theoretical models that simplify analysis and design. These systems operate under the principle that their output is directly proportional to their input within certain ranges. For instance, an amplifier in a control system behaves linearly as long as the input signal remains within a specific range. However, most physical systems exhibit inherent nonlinearity...
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...
Classification of Systems-I01:26

Classification of Systems-I

Linearity is a system property characterized by a direct input-output relationship, combining homogeneity and additivity.
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:

You might also read

Related Articles

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

Sort by
Same author

First Measurement of Time-Dependent CP Violation in the Flavor-Changing Neutral-Current Decay B^{0}→K_{S}^{0}μ^{+}μ^{-}.

Physical review letters·2026
Same author

Measurement of the Top-Quark Production Cross Section and Charge Asymmetry at LHCb.

Physical review letters·2026
Same author

First Evidence of the B_{s}^{0}→K^{-}π^{+}γ Decay.

Physical review letters·2026
Same author

Precision Measurement of CP Violation and Branching Fractions in B^{±}→K_{S}^{0}h^{±} (h=π, K) Decays and Search for the Rare Decay B_{c}^{±}→K_{S}^{0}K^{±}.

Physical review letters·2026
Same author

First Observation of the B[over ¯]_{s}^{0}→Λ_{c}^{+}Λ[over ¯]_{c}^{-} Decay and Evidence for the B[over ¯]^{0}→Λ_{c}^{+}Λ[over ¯]_{c}^{-} Decay.

Physical review letters·2026
Same author

Observation of the Rare Baryonic Decay B^{+}→pΛ[over ¯] and Measurement of its Weak Decay Parameter.

Physical review letters·2026

Related Experiment Videos

A hybrid linear/nonlinear training algorithm for feedforward neural networks.

S McLoone1, M D Brown, G Irwin

  • 1Advanced Control Engineering Research Centre, Department of Electrical and Electronic Engineering, The Queen's University of Belfast, Belfast BT9 5AH, UK.

IEEE Transactions on Neural Networks
|February 7, 2008
PubMed
Summary
This summary is machine-generated.

A novel hybrid optimization strategy enhances feedforward neural network training by integrating gradient-based and singular value decomposition (SVD) methods. This approach proves superior to traditional techniques, especially for complex Local Model Networks (LMNs).

Related Experiment Videos

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Science

Background:

  • Feedforward neural networks require efficient training algorithms.
  • Existing methods like second-order gradient descent have limitations.
  • Hybrid approaches offer potential for improved performance.

Purpose of the Study:

  • To introduce a new hybrid optimization strategy for training feedforward neural networks.
  • To evaluate the effectiveness of this strategy across different network architectures.
  • To demonstrate its advantages over conventional training methods.

Main Methods:

  • Combining gradient-based optimization for nonlinear weights.
  • Integrating singular value decomposition (SVD) for linear weights.
  • Applying the hybrid method to Multilayer Perceptrons (MLP), Radial Basis Function (RBF) networks, and Local Model Networks (LMNs).

Main Results:

  • The hybrid training scheme shows superior performance compared to second-order gradient methods.
  • The method is particularly effective for Local Model Networks (LMNs).
  • Demonstrated effectiveness in scenarios with a high linear to nonlinear parameter ratio.

Conclusions:

  • The proposed hybrid optimization strategy offers a significant advancement in neural network training.
  • This method provides a robust and efficient alternative for training complex feedforward architectures.
  • The hybrid approach is especially beneficial for the LMN architecture.