小样本元学习综述：A Concise Review of Recent Few-shot Meta-learning Methods

原文链接 小样本学习与智能前沿 。 在这个公众号后台回复'CRR-FMM'，即可获得电子资源。

1 Introduction

In this short communication, we present a concise review of recent representative meta- learning methods for few-shot image classification. We re- fer to such methods as few-shot meta-learning methods. Af- ter establishing necessary notation, we first mathematically formulate few-shot learning and offer a pseudo-coded algo- rithm for general few-shot training and evaluation. We then provide a taxonomy and a gentle review of recent few-shot meta-learning methods, to help researchers quickly grasp the state-of-the-art methods in this field. Finally we summa- rize some vital challenges to conclude this review with new prospects.

MindMap

完整的mindmap：https://note.youdao.com/ynoteshare1/index.html?id=3cf041f5616d34bb4677d03434531c92&type=note

2. The Framework of Few-shot Meta-learning

2.1. Notation and definitions

A classifier