Jove
Visualize
Contact Us

Related Concept Videos

Linear Approximation in Frequency Domain01:26

Linear Approximation in Frequency Domain

128
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....
128
Improving Translational Accuracy02:07

Improving Translational Accuracy

2.7K
2.7K
Linear Approximation in Time Domain01:21

Linear Approximation in Time Domain

120
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,...
120
Forced Transdifferentiation01:28

Forced Transdifferentiation

2.0K
Transdifferentiation, also known as lineage reprogramming, was first discovered by Selman and Kafatos in 1974 in silkmoths. They observed that the moths’ cuticle-producing cells transformed into salt-producing cells. Many such cases of natural transdifferentiation occur in organisms. In humans, pancreatic alpha cells can become beta cells. In newts, the loss of the eye’s lens causes the pigmented epithelial cells to transdifferentiate into the lens cells.
Artificial...
2.0K
Properties of DTFT I01:24

Properties of DTFT I

496
In signal processing, Discrete-Time Fourier Transforms (DTFTs) play a critical role in analyzing discrete-time signals in the frequency domain. Various properties of the DTFTs such as linearity, time-shifting, frequency-shifting, time reversal, conjugation, and time scaling help understand and manipulate these signals for different applications.
The linearity property of DTFTs is fundamental. If two discrete-time signals are multiplied by constants a and b respectively, and then combined to...
496
Properties of the z-Transform I01:17

Properties of the z-Transform I

281
The z-transform is a fundamental tool in digital signal processing, enabling the analysis of discrete-time systems through its various properties. It is an invaluable tool for analyzing discrete-time systems, offering a range of properties that simplify complex signal manipulations. One fundamental property is linearity. For any two discrete-time signals, the z-transform of their linear combination equals the same linear combination of their individual z-transforms. This property is essential...
281

You might also read

Related Articles

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

Sort by
Same author

U-Trans: a foundation model for seismic waveform representation and enhanced downstream earthquake tasks.

Scientific reports·2026
Same author

Gabor-enhanced physics-informed neural networks for fast simulations of acoustic wavefields.

Neural networks : the official journal of the International Neural Network Society·2025
Same author

Physics-informed neural wavefields with Gabor basis functions.

Neural networks : the official journal of the International Neural Network Society·2024
Same author

Seismic evidence for a 1000 km mantle discontinuity under the Pacific.

Nature communications·2023
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: Aug 30, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

675

Direct domain adaptation through reciprocal linear transformations.

Tariq Alkhalifah1, Oleg Ovcharenko1

  • 1Physical Sciences and Engineering, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia.

Frontiers in Artificial Intelligence
|August 29, 2022
PubMed
Summary

This study introduces a direct domain adaptation (DDA) method to improve neural network training using synthetic and real-world data. The approach enhances feature alignment between domains, boosting model performance on new datasets.

Keywords:
MNIST classificationcovariate shiftdeep learningdomain adaptationsupervised learning

More Related Videos

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.2K

Related Experiment Videos

Last Updated: Aug 30, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

675
Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.2K

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Supervised neural networks often struggle with performance degradation when trained on synthetic data and applied to real-world data.
  • Domain adaptation techniques aim to bridge the gap between different data distributions, improving model generalization.

Purpose of the Study:

  • To propose a novel direct domain adaptation (DDA) method for enhancing neural network training on synthetic data using real-world data features.
  • To improve the alignment of features between source (synthetic) and target (real-world) domains without altering network architecture or training protocols.

Main Methods:

  • The DDA method applies linear operations: cross-correlation with sample pixels and convolution with the mean of autocorrelated images from the other domain.
  • During training, source data is processed using target domain statistics, and vice-versa during inference.
  • The method was applied to train a convolutional neural network on MNIST (source) and test on MNIST-M (target).

Main Results:

  • The DDA approach achieved 70% accuracy on the MNIST-M test dataset.
  • Principal Component Analysis (PCA) and t-distributed Stochastic Neighbor Embedding (t-SNE) visualizations demonstrated improved feature similarity between source and target domains after DDA.
  • The proposed transformations align input features from different domains effectively.

Conclusions:

  • The direct domain adaptation method successfully enhances the utility of synthetic data for training neural networks by incorporating real-world data features.
  • DDA offers a data-centric approach to domain adaptation, improving model performance and feature representation alignment.
  • This technique shows promise for improving generalization in machine learning models across different data distributions.