Few-Shot Learning With Global Class Representations (2019-08-14T00:00:00.000000Z)

TL;DR

This paper proposes to tackle the challenging few-shot learning (FSL) problem by learning global class representations using both base and novel class training samples, and an effective sample synthesis strategy is developed to avoid overfitting.

Abstract

In this paper, we propose to tackle the challenging few-shot learning (FSL) problem by learning global class representations using both base and novel class training samples. In each training episode, an episodic class mean computed from a support set is registered with the global representation via a registration module. This produces a registered global class representation for computing the classification loss using a query set. Though following a similar episodic training pipeline as existing meta learning based approaches, our method differs significantly in that novel class training samples are involved in the training from the beginning. To compensate for the lack of novel class training samples, an effective sample synthesis strategy is developed to avoid overfitting. Importantly, by joint base-novel class training, our approach can be easily extended to a more practical yet challenging FSL setting, i.e., generalized FSL, where the label space of test data is extended to both base and novel classes. Extensive experiments show that our approach is effective for both of the two FSL settings.

Authors

T. Xiang

18 papers

Weiran Huang

3 papers

Tiange Luo

1 papers

TL;DR

Abstract

Authors

References39 items

Large-Scale Few-Shot Learning: Knowledge Transfer With Class Hierarchy

Adaptive Posterior Learning: few-shot learning with a surprise-based memory module

MetaGAN: An Adversarial Approach to Few-Shot Learning

Meta-Learning with Latent Embedding Optimization

Meta-learning with differentiable closed-form solvers

Dynamic Few-Shot Visual Learning Without Forgetting

Memory Matching Networks for One-Shot Image Recognition

Reptile: a Scalable Metalearning Algorithm

On First-Order Meta-Learning Algorithms

Meta-Learning for Semi-Supervised Few-Shot Classification

Low-Shot Learning from Imaginary Data

Learning to Compare: Relation Network for Few-Shot Learning

Few-Shot Learning with Graph Neural Networks

Few-Shot Learning Through an Information Retrieval Lens

A Unified Approach for Conventional Zero-Shot, Generalized Zero-Shot, and Few-Shot Learning

Good Semi-supervised Learning That Requires a Bad GAN

Semantic Autoencoder for Zero-Shot Learning

Prototypical Networks for Few-shot Learning

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

Learning to Remember Rare Events

Meta Networks

Optimization as a Model for Few-Shot Learning

Learning feed-forward one-shot learners

Matching Networks for One Shot Learning

Towards a Neural Statistician

Predicting Visual Exemplars of Unseen Classes for Zero-Shot Learning

One-shot Learning with Memory-Augmented Neural Networks

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

One-Shot Generalization in Deep Generative Models

Synthesized Classifiers for Zero-Shot Learning

Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

Deep Residual Learning for Image Recognition

An embarrassingly simple approach to zero-shot learning

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Going deeper with convolutions

ImageNet Large Scale Visual Recognition Challenge

Backpropagation Applied to Handwritten Zip Code Recognition

Siamese Neural Networks for One-Shot Image Recognition

One shot learning of simple visual concepts

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names