Leveraging the Feature Distribution in Transfer-based Few-Shot Learning

Abstract

Few-shot classification is a challenging problem due to the uncertainty caused by using few labelled samples. In the past few years, many methods have been proposed to solve few-shot classification, among which transfer-based methods have proved to achieve the best performance. Following this vein, in this paper we propose a novel transfer-based method that builds on two successive steps: 1) preprocessing the feature vectors so that they become closer to Gaussian-like distributions, and 2) leveraging this preprocessing using an optimal-transport inspired algorithm (in the case of transductive settings). Using standardized vision benchmarks, we prove the ability of the proposed methodology to achieve state-of-the-art accuracy with various datasets, backbone architectures and few-shot settings.

References in corpus (19)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Wide Residual Networks
• Theoretical Models of Learning to Learn
• Matching Networks for One Shot Learning
• Meta-Learning for Semi-Supervised Few-Shot Classification
• Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learning
• Manifold Mixup: Better Representations by Interpolating Hidden States
• Few-Shot Learning with Graph Neural Networks
• SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learning
• Meta-Learning with Latent Embedding Optimization
• A Two-Stage Approach to Few-Shot Learning for Image Recognition
• Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions
• Meta-Learned Confidence for Few-shot Learning
• Image Deformation Meta-Networks for One-Shot Learning
• Prototype Rectification for Few-Shot Learning
• Are Few-Shot Learning Benchmarks too Simple ? Solving them without Task Supervision at Test-Time
• Graph-based Interpolation of Feature Vectors for Accurate Few-Shot Classification
• Generalized Adaptation for Few-Shot Learning

Cited by in corpus (10)

• Automated Human Cell Classification in Sparse Datasets using Few-Shot Learning
• Sill-Net: Feature Augmentation with Separated Illumination Representation
• Is Support Set Diversity Necessary for Meta-Learning?
• Few-Shot Learning by Integrating Spatial and Frequency Representation
• Ontology-based n-ball Concept Embeddings Informing Few-shot Image Classification
• ORBIT: A Real-World Few-Shot Dataset for Teachable Object Recognition
• How to trust unlabeled data? Instance Credibility Inference for Few-Shot Learning
• Extended Few-Shot Learning: Exploiting Existing Resources for Novel Tasks
• A Strong Baseline for Semi-Supervised Incremental Few-Shot Learning
• Modular Adaptation for Cross-Domain Few-Shot Learning