Learning to Compare: Relation Network for Few-Shot Learning

TL;DR

The paper introduces the Relation Network, a simple and flexible deep learning framework that learns to compare images for few-shot and zero-shot classification, achieving strong results across multiple benchmarks.

Contribution

It proposes a novel end-to-end trainable relation network that learns a deep similarity metric for few-shot learning, unifying few-shot and zero-shot classification tasks.

Findings

Achieves state-of-the-art results on five benchmarks.

Effectively generalizes to zero-shot learning.

Provides a unified approach for few-shot and zero-shot classification.

Abstract

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.