Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Linear Approximation in Frequency Domain

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....

You might also read

Related Articles

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

Sort by

Same author

Hierarchical feature based dual contrastive multiview clustering.

Neural networks : the official journal of the International Neural Network Society·2026

Same author

Unsupervised feature selection via row-sparse local preserving projection.

Neural networks : the official journal of the International Neural Network Society·2026

Same author

A Unified Framework for Pseudo-Supervised Clustering via Weighted Sample Aggregation.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Projection with mixed-size anchor graphs.

Neural networks : the official journal of the International Neural Network Society·2026

Same author

SimMTC: Simple Multi-View Tensor Clustering.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same author

Unsupervised fine-tuning of vision-language models by fusing classifier tuning and visual prompt tuning.

Neural networks : the official journal of the International Neural Network Society·2026

Search research articles

Related Experiment Video

Updated: Sep 29, 2025

Assisted Selection of Biomarkers by Linear Discriminant Analysis Effect Size LEfSe in Microbiome Data

Assisted Selection of Biomarkers by Linear Discriminant Analysis Effect Size LEfSe in Microbiome Data

Published on: May 16, 2022

Fast Locality Discriminant Analysis With Adaptive Manifold Embedding.

Feiping Nie, Xiaowei Zhao, Rong Wang

IEEE Transactions on Pattern Analysis and Machine Intelligence

|March 25, 2022

Summary

Fast Locality Discriminant Analysis (FLDA) improves dimensionality reduction by using an anchor-based strategy to handle non-Gaussian data. This method efficiently captures data manifold structure and adaptively updates weights in a subspace, reducing noise and computation time.

More Related Videos

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Published on: July 1, 2014

Related Experiment Videos

Last Updated: Sep 29, 2025

Assisted Selection of Biomarkers by Linear Discriminant Analysis Effect Size LEfSe in Microbiome Data

Assisted Selection of Biomarkers by Linear Discriminant Analysis Effect Size LEfSe in Microbiome Data

Published on: May 16, 2022

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Published on: July 1, 2014

Area of Science:

Computer Science
Machine Learning
Data Mining

Background:

Linear Discriminant Analysis (LDA) is effective for dimensionality reduction but relies on global and local structure consistency.
Existing local LDA formulations have suboptimal learning schemes, pre-learning relationships in noisy original spaces, leading to high computational costs.

Purpose of the Study:

To propose a Fast Locality Discriminant Analysis (FLDA) framework to address the limitations of existing LDA methods.
To improve the efficiency and effectiveness of dimensionality reduction techniques.

Main Methods:

An anchor-based strategy is employed to divide non-Gaussian data classes into Gaussian-obeying sub-blocks.
The framework captures data manifold structure by learning fuzzy membership between data points and anchor points, reducing computation.
Weights are adaptively updated in a subspace, suppressing irrelevant information and noise from high-dimensional data.

Main Results:

FLDA demonstrates efficiency and effectiveness across toy, UCI benchmark, and imbalanced datasets.
The proposed method successfully handles non-Gaussian distributions and reduces computational complexity.
Adaptive weight updates in a suppressed subspace improve the handling of noisy and redundant features.

Conclusions:

FLDA offers a significant improvement over traditional LDA and its local variants.
The anchor-based strategy and adaptive subspace weighting contribute to enhanced performance and efficiency.
FLDA is a robust dimensionality reduction technique suitable for various data types and complexities.