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

Convolution: Math, Graphics, and Discrete Signals01:24

Convolution: Math, Graphics, and Discrete Signals

1.1K
In any LTI (Linear Time-Invariant) system, the convolution of two signals is denoted using a convolution operator, assuming all initial conditions are zero. The convolution integral can be divided into two parts: the zero-input or natural response and the zero-state or forced response, with t0 indicating the initial time.
To simplify the convolution integral, it is assumed that both the input signal and impulse response are zero for negative time values. The graphical convolution process...
1.1K
Convolution Properties I01:20

Convolution Properties I

680
Convolution computations can be simplified by utilizing their inherent properties.
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
680
Convolution Properties II01:17

Convolution Properties II

648
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...
648
Vector Representation of Complex Numbers01:16

Vector Representation of Complex Numbers

602
Complex numbers, represented in Cartesian coordinates, can also be visualized as vectors. These vectors can be expressed in polar form, emphasizing their magnitude and angle. When a complex number is input into a function, the output is another complex number, highlighting the function's zero point from which the vector representation can originate.
Consider a function defined as the product of the complex factors in the numerator divided by the product of the complex factors in the...
602
Complex Numbers01:29

Complex Numbers

489
The real number system cannot represent the square root of a negative number, which restricts solutions for certain equations, such as quadratics with negative discriminants. To address this, the complex number system was developed, introducing the imaginary unit i, where i = √(-1). This extension allows for the representation of all roots, including those involving negative radicands.A complex number is written in the form x + yi, where x and y are real numbers. Here, x represents the...
489
Deconvolution01:20

Deconvolution

683
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
683

You might also read

Related Articles

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

Sort by
Same author

Multi-modal AI for comprehensive breast cancer prognostication.

Nature communications·2026
Same author

To Compress or Not to Compress-Self-Supervised Learning and Information Theory: A Review.

Entropy (Basel, Switzerland)·2024
Same author

Navigating to objects in the real world.

Science robotics·2023
Same author

Catalyzing next-generation Artificial Intelligence through NeuroAI.

Nature communications·2023
Same author

Deep learning, reinforcement learning, and world models.

Neural networks : the official journal of the International Neural Network Society·2022
Same author

Randomized algorithms for distributed computation of principal component analysis and singular value decomposition.

Advances in computational mathematics·2021
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

Related Experiment Videos

A Mathematical Motivation for Complex-Valued Convolutional Networks.

Mark Tygert1, Joan Bruna2, Soumith Chintala3

  • 1tygert@fb.com.

Neural Computation
|February 19, 2016
PubMed
Summary
This summary is machine-generated.

Complex-valued convolutional networks (convnets) offer a direct link to wavelet analysis, enabling data-driven multiscale signal processing. This connection allows the application of rigorous mathematical tools from wavelet theory to these advanced neural networks.

Related Experiment Videos

Area of Science:

  • Computational Mathematics
  • Signal Processing
  • Machine Learning

Background:

  • Standard real-valued convolutional neural networks (convnets) lack a direct, rigorous connection to wavelet theory.
  • Complex-valued convnets process data through a recursive composition of convolution, absolute value, and local averaging.
  • Existing convnet architectures do not inherently map to the established mathematical frameworks of wavelets.

Purpose of the Study:

  • To establish an exact correspondence between complex-valued convnets and data-driven wavelet representations.
  • To demonstrate how complex-valued convnets can perform multiscale spectral analysis of real-valued random vectors.
  • To leverage the mathematical rigor of wavelet analysis for understanding complex-valued convnets.

Main Methods:

  • Utilizing a complex-valued convolutional network architecture involving convolution with complex-valued vectors.
  • Applying absolute value operations to intermediate results, followed by local averaging.
  • Configuring convnet filters as windowed complex-valued exponentials to achieve spectral analysis capabilities.

Main Results:

  • Complex-valued convnets can be interpreted as data-driven multiscale windowed power spectra, absolute spectra, multiwavelet absolute values, or nonlinear multiwavelet packets.
  • An exact correspondence is shown between complex-valued convnets and data-driven wavelets, unlike standard real-valued convnets.
  • This correspondence is particularly evident when convnet filters are designed as windowed complex-valued exponentials.

Conclusions:

  • Complex-valued convnets provide a direct and rigorous framework for data-driven multiscale signal processing, analogous to wavelets.
  • The established mathematical theory of wavelets can be directly applied to analyze complex-valued convnets.
  • This connection opens new avenues for advanced signal analysis and feature extraction using deep learning models.