SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learning

2019/11/12 by Yan Wang, Wei‐Lun Chao, Wang, Yan +5 · 29 citations
Computer Science · #Domain Adaptation and Few-Shot Learning #Multimodal Machine Learning Applications #Machine Learning and Data Classification

paper · pdf · doi:10.48550/arxiv.1911.04623

Abstract

Few-shot learners aim to recognize new object classes based on a small number of labeled training examples. To prevent overfitting, state-of-the-art few-shot learners use meta-learning on convolutional-network features and perform classification using a nearest-neighbor classifier. This paper studies the accuracy of nearest-neighbor baselines without meta-learning. Surprisingly, we find simple feature transformations suffice to obtain competitive few-shot learning accuracies. For example, we find that a nearest-neighbor classifier used in combination with mean-subtraction and L2-normalization outperforms prior results in three out of five settings on the miniImageNet dataset.

Citations

• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

Cited by

• Visual Species Recognition with Large Multimodal Models as Post-Hoc Correctors
• Decoupling Template Bias in CLIP: Harnessing Empty Prompts for Enhanced Few-Shot Learning
• Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions
• Leveraging the Feature Distribution in Transfer-based Few-Shot Learning
• A multimodal whole-slide foundation model for pathology
• Interventional Few-Shot Learning
• Enhancing Few-Shot Classification of Benchmark and Disaster Imagery with ABHFA-Net
• Revisiting Pose-Normalization for Fine-Grained Few-Shot Recognition
• Identifying and Compensating for Feature Deviation in Imbalanced Deep Learning
• MOCHA: Multi-modal Objects-aware Cross-arcHitecture Alignment
• Is Meta-Learning Out? Rethinking Unsupervised Few-Shot Classification with Limited Entropy
• Self-training for Few-shot Transfer Across Extreme Task Differences
• On Episodes, Prototypical Networks, and Few-shot Learning
• Data-Efficient Point Cloud Semantic Segmentation Pipeline for Unimproved Roads
• Separating Knowledge and Perception with Procedural Data
• Transductive Information Maximization For Few-Shot Learning
• Foundation Models for Bioacoustics -- a Comparative Review
• Gaussian Process Meta Few-shot Classifier Learning via Linear Discriminant Laplace Approximation
• Few-shot Intent Classification and Slot Filling with Retrieved Examples
• DecisiveNets: Training Deep Associative Memories to Solve Complex Machine Learning Problems
• Fuzzy Simplicial Networks: A Topology-Inspired Model to Improve Task Generalization in Few-shot Learning
• Learning by Examples Based on Multi-level Optimization
• Model-Agnostic Graph Regularization for Few-Shot Learning
• Towards Enabling Meta-Learning from Target Models
• Learning Adaptive Classifiers Synthesis for Generalized Few-Shot Learning
• PictSure: Pretraining Embeddings Matters for In-Context Learning Image Classifiers
• Handcrafted Histological Transformer (H2T): Unsupervised representation of whole slide images
• Structure-Enhanced Meta-Learning For Few-Shot Graph Classification
• Modeling and Optimization Trade-off in Meta-learning
• Overcoming Data Scarcity in Scanning Tunnelling Microscopy Image Segmentation
• BioCLIP 2: Emergent Properties from Scaling Hierarchical Contrastive Learning

Related

• A Closer Look at Few-shot Classification
• P3DC-Shot: Prior-Driven Discrete Data Calibration for Nearest-Neighbor Few-Shot Classification
• Few-Shot Learning with Localization in Realistic Settings
• Few-shot Learning via Dependency Maximization and Instance Discriminant Analysis
• Few-shot Image Classification: Just Use a Library of Pre-trained Feature Extractors and a Simple Classifier