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 Properties I01:20

Convolution Properties I

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:
Convolution: Math, Graphics, and Discrete Signals01:24

Convolution: Math, Graphics, and Discrete Signals

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...
Convolution Properties II01:17

Convolution Properties II

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...
Second Derivatives and Laplace Operator01:22

Second Derivatives and Laplace Operator

The first order operators using the del operator include the gradient, divergence and curl. Certain combinations of first order operators on a scalar or vector function yield second order expressions. Second-order expressions play a very important role in mathematics and physics. Some second order expressions include the divergence and curl of a gradient function, the divergence and curl of a curl function, and the gradient of a divergence function.
Consider a scalar function. The curl of its...
Partial Differential Equations01:21

Partial Differential Equations

A stone dropped into a still pond generates waves that propagate outward in circular patterns, creating a dynamic surface whose elevation depends on both position and time. At any given location, the water level oscillates as the wave passes, while at any fixed moment, the surface exhibits smooth, curved structures extending across space. This dual dependence requires a mathematical description that accounts for variation in multiple variables simultaneously.At a fixed point on the water...
Region of Convergence of Laplace Tarnsform01:20

Region of Convergence of Laplace Tarnsform

The Region of Convergence (ROC) is a fundamental concept in signal processing and system analysis, particularly associated with the Laplace transform. The ROC represents an area in the complex plane where the Laplace transform of a given signal converges, determining the transform's applicability and utility.
Consider a decaying exponential signal that begins at a specific time. When deriving its Laplace transform, the time-domain variable is replaced with a complex variable. This substitution...

You might also read

Related Articles

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

Sort by
Same author

Osimertinib with chemotherapy in advanced non-small cell lung cancer with vaginal metastasis harboring <i>EGFR</i> exon 21 L858R and <i>NTRK1</i> exon 16 V679M mutations: a case report with prolonged progression-free survival.

Gynecology and pelvic medicine·2026
Same author

Shexiang Tongxin Dropping Pills targeting Piezo1/Ca<sup>2+</sup>/NLRP3 axis attenuates vascular endothelial inflammation.

Chinese medicine·2026
Same author

Cardiovascular toxicity of ethyl maltol exposure: a comprehensive investigation from virtual screening to experimental validation.

Archives of toxicology·2026
Same author

Intrahepatic cholestasis of pregnancy and offspring neurodevelopment: bile acid-mediated mechanisms and long-term neurodevelopmental outcomes.

Archives of gynecology and obstetrics·2026
Same author

A retrospective study on risk factors, diagnostic approaches, and pregnancy outcomes in patients with umbilical artery and vein thrombosis.

International journal of gynaecology and obstetrics: the official organ of the International Federation of Gynaecology and Obstetrics·2026
Same author

Exosomal novel-miRNA-126 mediates vascular endothelial dysfunction by targeting AhR-NLRP3 pathway in nonalcoholic steatohepatitis.

Scientific reports·2025

Related Experiment Video

Updated: Jul 3, 2026

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Physics-encoded convolutional neural operators for parametric PDEs: A convergence-guaranteed framework via

Yu Liu1, Yanfei Chen1, Ruihao Liu1

  • 1National Engineering Laboratory for Pipeline Safety, China University of Petroleum (Beijing), MOE Key Laboratory of Petroleum Engineering, Beijing, 102249, China.

Neural Networks : the Official Journal of the International Neural Network Society
|July 1, 2026
PubMed
Summary

Convolutional Neural Operators with Physics-Encoded Kernels (CNO-PEK) offer provable approximation bounds for parametric PDEs without training data. This novel framework achieves significant error reduction and computational speedup, bridging numerical analysis and deep learning.

Keywords:
Convergence theoryOperator learningParametric PDEsPhysics-encoded neural networksScientific machine learningZero-shot generalization

Related Experiment Videos

Last Updated: Jul 3, 2026

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Area of Science:

  • Numerical analysis and scientific machine learning.
  • Development of novel operator learning frameworks for partial differential equations (PDEs).

Background:

  • Existing operator learning methods like PINNs, DeepONet, and FNO face limitations.
  • PINNs lack rigorous convergence guarantees, while DeepONet and FNO require extensive training data.
  • A gap exists for methods with provable approximation bounds, zero training data requirements, and fast inference.

Purpose of the Study:

  • To introduce Convolutional Neural Operators with Physics-Encoded Kernels (CNO-PEK) to address limitations in current operator learning methods.
  • To develop a framework with provable O(hk+1-m) approximation-capacity bounds, zero training data needs, and efficient multi-query inference.

Main Methods:

  • CNO-PEK encodes physics directly into the network architecture by pre-computing differential operators as spatially-adaptive convolutional kernels via the Physics-Informed Kernel Field (PIKF).
  • Kernels are analytically derived from polynomial consistency conditions, bypassing the need for automatic differentiation or solution snapshots.
  • The framework leverages neural network approximation capabilities while maintaining theoretical guarantees.

Main Results:

  • Empirical validation confirmed O(hk+1-m) approximation-capacity bounds.
  • Achieved a 60.9% systematic error reduction compared to linear finite elements (1.72% vs. 4.40% mean error) without training data.
  • Demonstrated physics-driven zero-shot parameter generalization across 1000 unseen parameter values with <3% error.
  • Showcased computational speedups of 6.9-13.6x for multi-query scenarios.

Conclusions:

  • CNO-PEK provides a theoretically grounded operator learning framework that merges classical numerical analysis with deep learning.
  • The method is highly suitable for parametric studies, uncertainty quantification, and real-time applications demanding data efficiency and approximation guarantees.
  • CNO-PEK effectively bridges the gap between data-driven and physics-informed approaches in operator learning.