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

Linearization and Approximation01:26

Linearization and Approximation

115
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...
115
Second Derivatives of Implicit Functions01:29

Second Derivatives of Implicit Functions

108
Elliptical arches are fundamental in architectural and structural engineering, offering aesthetic appeal and structural efficiency. The shape of an elliptical arch follows a constrained geometric relationship where the height and horizontal position are implicitly related. This means that the height y cannot be explicitly expressed as a function of the horizontal position x, necessitating implicit differentiation for slope and curvature analysis.The equation of an ellipse centered at the origin...
108
Convolution Properties II01:17

Convolution Properties II

627
The important convolution properties include width, area, differentiation, and integration properties.
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
627
Application of Linearization and Approximation01:29

Application of Linearization and Approximation

121
A drone flying through complex terrain often relies on more than one sensing method to estimate small changes in altitude. Along with direct measurements, air pressure provides a useful indirect indicator of vertical movement. Atmospheric pressure decreases as altitude increases, and this relationship is commonly described using an exponential model. Although accurate, converting pressure measurements into altitude values requires calculations that are too complex to perform repeatedly during...
121
Linear Approximation in Frequency Domain01:26

Linear Approximation in Frequency Domain

408
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....
408
Separable Differential Equations01:20

Separable Differential Equations

164
A separable differential equation is a type of first-order differential equation where the derivative dy/dx can be expressed as a product of two functions: one that depends only on x and another that depends only on y. This allows for the rearrangement of the equation so that all terms involving y are on one side, and all terms involving x are on the other. This process, known as the separation of variables, simplifies the process of solving the equation by enabling the integration of both...
164

You might also read

Related Articles

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

Sort by
Same author

Deep Kernel Principal Component Analysis for multi-level feature learning.

Neural networks : the official journal of the International Neural Network Society·2023
Same author

Learning With Asymmetric Kernels: Least Squares and Feature Interpretation.

IEEE transactions on pattern analysis and machine intelligence·2023
Same author

Disentangled Representation Learning and Generation With Manifold Optimization.

Neural computation·2022
Same author

Compressing Features for Learning With Noisy Labels.

IEEE transactions on neural networks and learning systems·2022
Same author

Towards a Unified Quadrature Framework for Large-Scale Kernel Machines.

IEEE transactions on pattern analysis and machine intelligence·2021
Same author

Unsupervised learning of disentangled representations in deep restricted kernel machines with orthogonality constraints.

Neural networks : the official journal of the International Neural Network Society·2021

Related Experiment Video

Updated: Mar 1, 2026

Quantifying Intermembrane Distances with Serial Image Dilations
07:45

Quantifying Intermembrane Distances with Serial Image Dilations

Published on: September 28, 2018

6.8K

Deep Restricted Kernel Machines Using Conjugate Feature Duality.

Johan A K Suykens1

  • 1KU Leuven ESAT-STADIUS, B-3001 Leuven, Belgium johan.suykens@esat.kuleuven.be.

Neural Computation
|June 1, 2017
PubMed
Summary
This summary is machine-generated.

This study introduces deep restricted kernel machines (RKMs), a novel framework unifying deep learning and kernel methods. Deep RKMs offer new foundations for machine learning by integrating techniques like support vector machines and kernel PCA.

More Related Videos

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

1.7K

Related Experiment Videos

Last Updated: Mar 1, 2026

Quantifying Intermembrane Distances with Serial Image Dilations
07:45

Quantifying Intermembrane Distances with Serial Image Dilations

Published on: September 28, 2018

6.8K
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

1.7K

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Computational Science

Background:

  • Deep learning models like Restricted Boltzmann Machines (RBMs) utilize visible and hidden units in a bipartite graph.
  • Kernel machines encompass methods such as least squares support vector machines (LS-SVMs), kernel PCA, and Parzen models.
  • Existing frameworks lack a unified approach connecting deep learning architectures with kernel machine principles.

Purpose of the Study:

  • To propose a novel theory of deep restricted kernel machines (RKMs).
  • To establish new foundations for deep learning by integrating kernel machine concepts.
  • To provide a unified framework for diverse machine learning techniques.

Main Methods:

  • Characterizing kernel machines using conjugate feature duality to establish a visible-hidden unit representation.
  • Relating this representation to the energy form of Restricted Boltzmann Machines in a non-probabilistic, continuous variable setting.
  • Developing deep RKMs by coupling individual RKMs, demonstrated with LS-SVM and kernel PCA levels.

Main Results:

  • Introduced Restricted Kernel Machine (RKM) representations where dual variables act as hidden features.
  • Demonstrated the coupling of RKMs to form deep RKM architectures.
  • Showcased the framework's flexibility by illustrating a three-level deep RKM and indicating potential for training deep feedforward neural networks.

Conclusions:

  • Deep RKMs offer a unified theoretical foundation for deep learning and kernel machines.
  • The proposed framework provides a novel perspective on integrating diverse machine learning methodologies.
  • This approach opens new avenues for developing advanced deep learning architectures with enhanced capabilities.