Learning to Compare: Relation Network for Few-Shot Learning

We present a conceptually simple, flexible, and general framework for few-shot learning, where a classifier must learn to recognise new classes given only few examples from each. Our method, called the Relation Network (RN), is trained end-to-end from scratch. During meta-learning, it learns to learn a deep distance metric to compare a small number of images within episodes, each of which is designed to simulate the few-shot setting. Once trained, a RN is able to classify images of new classes by computing relation scores between query images and the few examples of each new class without further updating the network. Besides providing improved performance on few-shot learning, our framework is easily extended to zero-shot learning. Extensive experiments on five benchmarks demonstrate that our simple approach provides a unified and effective approach for both of these two tasks.

• Publication year

  2017

• Venue

  2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

• Publication date

  2017-11-16

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/CVPR.2018.00131arXiv 1711.06025
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

  2017influential reference
• Meta Networks

  2017influential reference
• Prototypical Networks for Few-shot Learning

  2017influential reference
• Learning to Remember Rare Events

  2017influential reference
• Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly

  2017influential reference
• Meta-Learning with Temporal Convolutions

  2017influential reference
• A simple neural network module for relational reasoning

  2017cited by this paper
• Semantic Autoencoder for Zero-Shot Learning

  2017cited by this paper
• Zero-Shot Recognition via Structured Prediction

  2016cited by this paper
• Semi-supervised Vocabulary-Informed Learning

  2016cited by this paper
• Towards a Neural Statistician

  2016cited by this paper
• Learning a Deep Embedding Model for Zero-Shot Learning

  2016influential reference
• Meta-Learning with Memory-Augmented Neural Networks

  2016influential reference
• An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild

  2016cited by this paper
• Matching Networks for One Shot Learning

  2016cited by this paper
• Learning Deep Representations of Fine-Grained Visual Descriptions

  2016cited by this paper
• Synthesized Classifiers for Zero-Shot Learning

  2016cited by this paper
• Optimization as a Model for Few-Shot Learning

  2016influential reference
• Latent Embeddings for Zero-Shot Classification

  2016cited by this paper
• Improving Semantic Embedding Consistency by Metric Learning for Zero-Shot Classiffication

  2016cited by this paper
• Learning feed-forward one-shot learners

  2016cited by this paper
• Zero-Shot Learning via Joint Latent Similarity Embedding

  2015cited by this paper
• Zero-Shot Learning via Semantic Similarity Embedding

  2015cited by this paper
• Label-Embedding for Image Classification

  2015cited by this paper
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

  2015cited by this paper
• An embarrassingly simple approach to zero-shot learning

  2015cited by this paper
• Siamese Neural Networks for One-Shot Image Recognition

  2015influential reference
• Predicting Deep Zero-Shot Convolutional Neural Networks Using Textual Descriptions

  2015cited by this paper
• MatchNet: Unifying feature and metric learning for patch-based matching

  2015cited by this paper
• Deep Residual Learning for Image Recognition

  2015cited by this paper
• Learning to compare image patches via convolutional neural networks

  2015cited by this paper
• Evaluation of output embeddings for fine-grained image classification

  2014cited by this paper
• A Unified Perspective on Multi-Domain and Multi-Task Learning

  2014cited by this paper
• Very Deep Convolutional Networks for Large-Scale Image Recognition

  2014cited by this paper
• Going deeper with convolutions

  2014cited by this paper
• Transductive Multi-view Embedding for Zero-Shot Recognition and Annotation

  2014cited by this paper
• Adam: A Method for Stochastic Optimization

  2014influential reference
• Zero-Shot Learning by Convex Combination of Semantic Embeddings

  2013cited by this paper
• DeViSE: A Deep Visual-Semantic Embedding Model

  2013cited by this paper
• OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks

  2013cited by this paper
• Zero-Shot Learning Through Cross-Modal Transfer

  2013cited by this paper
• Bayesian Face Revisited: A Joint Formulation

  2012cited by this paper
• Metric Learning for Large Scale Image Classification: Generalizing to New Classes at Near-Zero Cost

  2012cited by this paper
• ImageNet classification with deep convolutional neural networks

  2012cited by this paper
• Multiclass recognition and part localization with humans in the loop

  2011cited by this paper
• One shot learning of simple visual concepts

  2011influential reference
• One-shot learning of object categories

  2006influential reference
• Attribute-Based Classification for Zero-Shot Visual Object Categorization

  year unknowninfluential reference

• Complex-Valued Source-Free Domain Adaptation for PolSAR Image Classification

  2026cites this paper
• Dual-Branch Prototype Enhancement Network for Few-Shot Hyperspectral Image Classification

  2026cites this paper
• Few-shot with prototype augmentation for low-resource domains rumor detection

  2026cites this paper
• SCAFFNet: A Novel Object Classification Algorithm in Low-Altitude Scenarios for eVTOL

  2026cites this paper
• From Physics to Machine Learning and Back: Part II - Learning and Observational Bias in Prognostics and Health Management (PHM)

  2026cites this paper
• Semantic Modulated Prompting for Few-Shot Audio-Visual Classification

  2026cites this paper
• Few-Shot Cross-Domain Fault Diagnosis of Complex Mechanical Equipment Based on Multiangle Taylor Metric

  2026cites this paper
• Dynamic meta-learning with generative augmentation for cross-lingual Japanese few-shot named entity recognition

  2026cites this paper
• Feature-Space Generative Models for One-Shot Class-Incremental Learning

  2026cites this paper
• Prior knowledge-informed multi-task collaborative learning for few-shot fault diagnosis of aero-engines

  2026cites this paper
• Adaptive feature enhancement and distribution smoothing for improved few-shot image classification

  2026cites this paper
• Few-shot coral recognition via prototype refinement and segmentation-guided feature enhancement

  2026cites this paper
• Cross-Domain Few-Shot Learning for Hyperspectral Image Classification Based on Mixup Foundation Model

  2026cites this paper
• Meta-learning for few-shot open task recognition

  2026cites this paper
• Disentangling Domain-Invariant Features From Few Samples for Automatic Modulation Classification

  2026cites this paper
• A self-supervised framework for improved generalisability in ultrasound B-mode image segmentation

  2026cites this paper
• DMASG: A denoising memory adaptive spatial guidance network for soft sensor modelling in complex manufacturing processes

  2026cites this paper
• FALCON: Few-Shot Adversarial Learning for Cross-Domain Medical Image Segmentation

  2026cites this paper
• Efficient group attentive learning for few-shot image classification

  2026cites this paper
• Self-supervised transfer learning for few-shot classification on marine plankton images

  2026cites this paper
• 1S-DAug: One-Shot Data Augmentation for Robust Few-Shot Generalization

  2026cites this paper
• Improving Episodic Few-shot Visual Question Answering via Spatial and Frequency Domain Dual-calibration

  2026cites this paper
• Leveraging contrastive learning for cross-modal(X-modal) person identification

  2026cites this paper
• Towards few-shot malware classification with fine-grained and pattern-aware multi-prototype network

  2026cites this paper
• A Visually Explainable Dynamic Similarity Network for Few-Shot Classification

  2026cites this paper
• Few-Shot Semantic Segmentation in Remote Sensing: A Review on Definitions, Methods, Datasets, Advances and Future Trends

  2026cites this paper
• AdaMixCon: Difficulty-Aware Feature Enhancement for Few-Shot Hyperspectral Image Classification

  2026cites this paper
• Self-Regressive Prototype Refinement: Stepping from Local to Global Prototypes in Few-Shot Image Classification

  2026cites this paper
• BayPrAnoMeta: Bayesian Proto-MAML for Few-Shot Industrial Image Anomaly Detection

  2026cites this paper
• Correlation guided multi-teacher distillation for lightweight image retrieval

  2026cites this paper
• Unbiased Spatial-Temporal Atomic Fusion-based Zero-Shot Action Recognition

  2026cites this paper
• Leveraging CVAE Encoding for Backdoor Attacks in Few-Shot Learning With Prototypical Networks

  2026cites this paper
• Few-Shot Cross-Domain Fault Diagnosis of Transportation Motor Bearings Using MAML-GA

  2026cites this paper
• Retrospective Prototype Network Based on Center Difference Measure for Cross-Machine Few-Shot Fault Diagnosis

  2026cites this paper
• CrossHypergraph: Consistent High-Order Semantic Network for Few-Shot Image Classification

  2026cites this paper
• Few-shot transformers for image classification using masked self-distillation and optimal tokens based global-local feature interactions

  2026cites this paper
• Mitigating bias in Few Shot Class Incremental Learning with Feature Augmentation and Logits Mix-up

  2026cites this paper
• A Survey on Edge-Aware Graph Learning Methods

  2026cites this paper
• Exploiting Class-agnostic Visual Prior for Few-shot Keypoint Detection

  2026influential citation
• A novel multi-sensor fusion method for diagnosing insulation defects in gas-insulated substations guided by adaptive-attention and contrastive-based few-shot learning

  2026cites this paper
• IPEC: Test-Time Incremental Prototype Enhancement Classifier for Few-Shot Learning

  2026cites this paper
• Self-Supervised Few-Shot Modulation Recognition With Adversarial Representation Learning and Mixture of Experts

  2026cites this paper
• An In-Context Foundation Model for Predictive Process Monitoring on Event Logs

  2026cites this paper
• Prototypical aggregate network - boosting few-shot learning for medical image classification

  2026influential citation
• MACo-Net: a multi-domain adaptive collaborative network for few-shot fault diagnosis

  2026cites this paper
• ATARS: Adaptive task-Aware feature learning for Few-Shot Fine-Grained classification

  2026cites this paper
• A Text-Guided Prototype Feature Modulation Method for Few-Shot Semantic Segmentation

  2026cites this paper
• A novel few-shot meta-learning strategy for fault diagnosis of wastewater treatment process.

  2026cites this paper
• Cross few-shot learning-based query adaptive network for medical image segmentation

  2026cites this paper
• Spatio-temporal Decoupled Knowledge Compensator for Few-Shot Action Recognition.

  2026cites this paper
• CIFE-FSOD: Few-shot object detection via joint extraction of common and instance-specific features

  2026cites this paper
• Sparse Gaussian–Markov Modeling for Robust and Trustworthy Unknown Cyber Defense

  2026cites this paper
• ROSER: Few-Shot Robotic Sequence Retrieval for Scalable Robot Learning

  2026cites this paper
• Texture and shape enhanced feature reconstruction network for few-shot fine-grained aircraft classification

  2026cites this paper
• Knowledge-data-model-driven multimodal few-shot learning for hyperspectral fine classification: Generalization across sensor, category and scene

  2026cites this paper
• Multi-scale metric learning neural network with adaptive graphs for few-shot gas path fault diagnosis of variable cycle engine

  2026cites this paper
• A welding penetration prediction model for laser welding process based on self-supervised learning using physics-informed neural networks

  2026cites this paper
• Gaussian embedding metric learning for few-shot knowledge graph completion

  2026cites this paper
• Learning discriminative prototypes: Adaptive relation-aware refinement and patch-level contextual feature reweighting for few-shot classification.

  2026cites this paper
• Multiscale Classifier Ensemble Knowledge Distillation for Remote Sensing Image Classification

  2026cites this paper
• Consistency-Regularized GAN for Few-Shot SAR Target Recognition

  2026cites this paper
• PMCE: Probabilistic Multi-Granularity Semantics with Caption-Guided Enhancement for Few-Shot Learning

  2026influential citation
• Few-Shot Automatic Modulation Classification via Semi-Supervised Metric Learning and Lightweight Conv-Transformer Model

  2026cites this paper
• Toward Effective and Sustainable Traffic Management With Network Scaling Configuration: A Federated FSL Approach

  2026cites this paper
• Global and Local Attention-Based Multiscale Prototypical Network for Few-Shot Learning

  2026influential citation
• A 40-nm Training-Inference STT-MRAM Near-Memory Computing Macro for Memory-Augmented Neural Network Acceleration

  2026cites this paper
• Temporal Consistency and Variation-Guided Spatio-Temporal Aggregation for Few-Shot Action Recognition

  2026cites this paper
• FCGNN: Fuzzy Cognitive Graph Neural Networks With Concept Evolution for Few-Shot Learning

  2026cites this paper
• Robust Few-Shot segmentation for multi-modal medical images using Fourier-Guided style fusion and prototype alignment

  2026cites this paper
• Few-Shot Recognition Method of MFR Operation Modes via Graph-Based Transductive Prototype Propagation Network

  2026cites this paper
• I-FCSAM: An integrated framework of few-shot learning and segment anything model for vision-based indoor built environment management

  2026cites this paper
• A novel knowledge-informed quadratic neurons residual network for explainable fault diagnosis in few-shot scenarios

  2026cites this paper
• Debiased prototype network with relaxed contrastive distillation strategy for rotating machinery few-shot domain incremental fault diagnosis

  2026cites this paper
• Enhancing Few-Shot Surface Defect Recognition via Pre-Trained Large Generative Models

  2026influential citation
• A regularized deep self-expression feature augmentation network for few-shot unconstrained palmprint recognition

  2026cites this paper
• A Semisupervised Prototypical Network With Dynamic Threshold Pseudolabeling for Forest Classification

  2026cites this paper
• SOFA: Service-Oriented Fine-Grained Attack Traffic Detection With Meta Learning

  2026cites this paper
• Expert-Guided Explainable Few-Shot Learning with Active Sample Selection for Medical Image Analysis

  2026cites this paper
• Reconstruction Guided Few-shot Network For Remote Sensing Image Classification

  2026cites this paper
• EfficientFSL: Enhancing Few-Shot Classification via Query-Only Tuning in Vision Transformers

  2026cites this paper
• HCNM: Hierarchical cognitive neural model for small-sample image classification

  2025cites this paper
• Few-Shot SAR ATR via Multilevel Contrastive Learning and Dependence Matrix-Based Measurement

  2025influential citation
• A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techniques, Datasets, and Future Research Prospects

  2025cites this paper
• HILGEN: Hierarchically-Informed Data Generation for Biomedical NER Using Knowledgebases and Large Language Models

  2025cites this paper
• Automatically learns task-specific features for high-performance image classification

  2025cites this paper
• Dual-branch aggregation and edge refinement network for few shot semantic segmentation

  2025cites this paper
• Brain Inspired Adaptive Memory Dual-Net for Few-Shot Image Classification

  2025cites this paper
• Proto-Att-FSL: enhanced prototypical network for cross-domain few-shot airline sentiment classification

  2025cites this paper
• Deep Comparison Convolutional Neural Network for Industrial Localization

  2025cites this paper
• Few-Shot Underwater Acoustic Target Recognition Using Domain Adaptation and Knowledge Distillation

  2025cites this paper
• Mathematics-Inspired Models: A Green and Interpretable Learning Paradigm for Multimedia Computing

  2025cites this paper
• Enhanced few-shot object detection for remote sensing images based on target characteristics

  2025cites this paper
• SVD-Based Feature Reconstruction Metric Network With Active Contrast Loss for Few-Shot SAR Target Recognition

  2025cites this paper
• Deep Fourier-Based Task-Aware Metric Network for Few-Shot SAR Target Classification

  2025cites this paper
• FBINet: Few-Shot Semantic Segmentation With Foreground and Background Iteration

  2025cites this paper
• An Adaptive Hierarchical Framework With Contrastive Aggregation for Traffic Sign Classification

  2025cites this paper
• A Spatially Aware Few-Shot Approach to Classification of Radar Sounder Data

  2025cites this paper
• A robust transductive distribution calibration method for few-shot learning

  2025cites this paper
• Rethinking Semantic Segmentation With Multi-Grained Logical Prototype

  2025cites this paper
• Hard-Positive Prototypical Networks for Few-Shot Classification

  2025cites this paper