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

You might also read

Related Articles

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

Sort by

Same author

Thickness-Dependent Decoupling Charge Transport and NH 3 Sensing in Multilayer MoS 2 Transistors.

Nanotechnology·2026

Same author

Thermally Regulated Curing-Degradation Windows of Epoxidized Soybean Oil-Based Epoxy-Anhydride Liquid Plugs for Sustainable High-Temperature Sealing.

Molecules (Basel, Switzerland)·2026

Same author

Increased ULBP1 by doxorubicin sensitizes neuroblastoma to γδT-cell cytotoxicity.

Journal of immunology (Baltimore, Md. : 1950)·2026

Same author

Cytokine profiling reveals distinct inflammatory clusters and clinical correlations in antiphospholipid syndrome.

Clinical rheumatology·2026

Same author

Butyrate blocks cell cycle progression in colorectal cancer organoids partially through HDAC2 inhibition.

Frontiers in immunology·2026

Same author

Fracability evaluation of deep continental shale oil reservoirs based on combination weighting with game theory: a case study in the Lower Permian Fengcheng Formation in Mahu Sag, Junggar Basin.

Scientific reports·2026

Same journal

DARUMA: a gateway to fast and easy prediction of intrinsically disordered regions.

PeerJ. Computer science·2026

Same journal

Alzheimer's disease detection using a quantum deep neural network with Haralick feature extraction and simulated annealing optimization.

PeerJ. Computer science·2026

Same journal

Network anomaly detection using Deep Autoencoder and parallel Artificial Bee Colony algorithm-trained neural network.

PeerJ. Computer science·2026

Same journal

An anomaly detection model for multivariate time series with anomaly perception.

PeerJ. Computer science·2026

Same journal

Retraction: A wormhole attack detection method for tactical wireless sensor networks.

PeerJ. Computer science·2026

Same journal

Evaluation of mental disorder with prioritization of its type by utilizing the bipolar complex fuzzy decision-making approach based on Schweizer-Sklar prioritized aggregation operators.

PeerJ. Computer science·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Oct 24, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Research on image classification method based on improved multi-scale relational network.

Wenfeng Zheng¹, Xiangjun Liu¹, Lirong Yin²

¹School of Automation, University of Electronic Science and Technology of China, Chengdu, China.

Peerj. Computer Science

|August 16, 2021

Summary

This summary is machine-generated.

This study introduces a novel multi-scale meta-relational network for small sample learning in image classification. The approach enhances generalization ability and classification accuracy without requiring model fine-tuning.

Keywords:

Image classification Less sample learning META-SGD Meta-learning Model-independent Multi-scale characteristics Multi-scale relational network

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

Related Experiment Videos

Last Updated: Oct 24, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

Area of Science:

Computer Science
Artificial Intelligence
Machine Learning

Background:

Small sample learning is critical for deep learning, enabling models to learn from limited data.
Meta-learning, or "learning to learn," provides a framework to address small sample learning challenges by leveraging prior experience.
Image classification is a key application area where efficient learning from few examples is highly desirable.

Purpose of the Study:

To investigate how meta-learning can accelerate the learning process from a minimal number of sample images in image classification tasks.
To develop an improved meta-learning approach that addresses data distribution differences and optimizes initial feature representation.
To enhance the generalization performance of measurement learning in the context of small sample image classification.

Main Methods:

A multi-scale meta-relational network was designed, incorporating model-independent meta-learning algorithms.
The META-SGD approach was adapted, treating the inner learning rate as a learnable vector alongside model parameters.
Meta-training utilized model-independent meta-learning to identify optimal model parameters, while meta-validation and meta-testing omitted inner gradient iterations.

Main Results:

The proposed multi-scale meta-relational network demonstrated enhanced generalization ability for learned measurements.
The method significantly improved classification accuracy on benchmark datasets.
The approach successfully avoided the need for fine-tuning, a common requirement in model-independent meta-learning algorithms.

Conclusions:

The multi-scale meta-relational network offers a robust solution for small sample image classification.
This meta-learning strategy effectively improves model generalization and classification performance.
The developed method presents an efficient alternative to traditional fine-tuning techniques in meta-learning applications.