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

119
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...
119
Tangent Line01:26

Tangent Line

633
In differential calculus, understanding how a quantity changes at an exact point is central to interpreting dynamic systems. This can be illustrated by analyzing a car traveling along a winding road. The car’s trajectory is represented as a continuous curve, and the direction in which it moves at any instant is given by the tangent to that curve. In contrast, the secant line, intersecting the curve at two points, captures how the car’s position changes over an interval — an...
633
Neural Regulation01:37

Neural Regulation

43.9K
Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.
43.9K
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving01:29

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

376
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
376
Implicit Differentiation01:25

Implicit Differentiation

93
In classical mechanics, motion is often described through relationships between spatial coordinates and time. A car moving along a straight highway with constant acceleration serves as a simple case where velocity is an explicit function of time. This scenario results in a linear equation, enabling straightforward analysis using basic differentiation techniques.In contrast, a satellite in circular orbit follows a path defined by an implicit function. The position of the satellite is constrained...
93
Tangent to a Curve01:30

Tangent to a Curve

397
The graph of a function where each output is the square of the input creates a smooth curve that bends upward, becoming steeper as one moves further from the center. At any chosen position along this curve, the curve reaches a certain height depending on the input value. This position can be a reference for analyzing how the curve behaves in its immediate vicinity.To understand the change in the curve near a particular position, imagine selecting another point slightly ahead along the curve.
397

You might also read

Related Articles

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

Sort by
Same author

Exploring the Stochastic Regularisation in Normalisation Layers for Semi-Supervised Learning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

Embodied Spatial Affordance: Spatial-Aware Affordance Learning for Embodied Navigation and Manipulation.

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

Paving the Way for Point Cloud Video Representation Learning Using a PDE Model.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

Carbon-Nitrogen Metabolism Associated with Appearance Quality in Superior and Inferior Grains of Soft and Non-Soft Japonica Rice in Southern China.

Plants (Basel, Switzerland)·2026
Same author

ConsDreamer: Advancing Multi-View Consistency for Zero-Shot Text-to-3D Generation.

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

<i>In situ</i> synthesis of self-standing SbBi-porous carbon fibers enabling ultra-stable sodium-ion storage.

Chemical communications (Cambridge, England)·2026
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

Related Experiment Video

Updated: Mar 8, 2026

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

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

10.1K

Parameter-Efficient Fine-Tuning for Continual Learning: A Neural Tangent Kernel Perspective.

Jingren Liu, Zhong Ji, YunLong Yu

    IEEE Transactions on Pattern Analysis and Machine Intelligence
    |March 6, 2026
    PubMed
    Summary
    This summary is machine-generated.

    Parameter-efficient fine-tuning for continual learning (PEFT-CL) shows promise but lacks understanding. A new framework, NTK-CL, uses Neural Tangent Kernel theory to improve PEFT-CL performance by analyzing generalization gaps and feature orthogonality.

    Related Experiment Videos

    Last Updated: Mar 8, 2026

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

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    10.1K

    Area of Science:

    • Artificial Intelligence
    • Machine Learning
    • Deep Learning

    Background:

    • Parameter-efficient fine-tuning for continual learning (PEFT-CL) adapts models to new tasks while preventing catastrophic forgetting.
    • Mechanisms governing PEFT-CL performance and forgetting remain poorly understood.

    Purpose of the Study:

    • To analyze PEFT-CL dynamics using Neural Tangent Kernel (NTK) theory.
    • To identify key factors influencing generalization gaps and PEFT-CL performance.
    • To develop a novel framework, NTK-CL, for improved continual learning.

    Main Methods:

    • Utilized NTK theory to analyze PEFT-CL generalization gaps during training.
    • Identified training sample size, task feature orthogonality, and regularization as key factors.
    • Introduced NTK-CL framework with adaptive feature generation and task orthogonality constraints.

    Main Results:

    • NTK-CL triples sample feature representation, reducing task-interplay and generalization gaps.
    • Framework maintains intra-task NTK forms while attenuating inter-task NTK forms.
    • NTK-CL achieved state-of-the-art performance on PEFT-CL benchmarks.

    Conclusions:

    • NTK-CL provides a theoretical foundation for understanding and enhancing PEFT-CL.
    • Highlights the interplay between feature representation, task orthogonality, and generalization.
    • Contributes to developing more efficient continual learning systems.