Reinforced Attention for Few-Shot Learning and Beyond

TL;DR

This paper introduces a reinforcement learning-based attention mechanism for few-shot learning, enabling the model to adaptively focus on relevant image regions, resulting in improved generalization and discriminative representations.

Contribution

It proposes a novel reinforcement learning approach to train an attention agent that localizes important regions, enhancing few-shot learning performance.

Findings

Reinforced attention improves few-shot classification accuracy.

The method generalizes well to unseen classes.

It also benefits standard image classification tasks.

Abstract

Few-shot learning aims to correctly recognize query samples from unseen classes given a limited number of support samples, often by relying on global embeddings of images. In this paper, we propose to equip the backbone network with an attention agent, which is trained by reinforcement learning. The policy gradient algorithm is employed to train the agent towards adaptively localizing the representative regions on feature maps over time. We further design a reward function based on the prediction of the held-out data, thus helping the attention mechanism to generalize better across the unseen classes. The extensive experiments show, with the help of the reinforced attention, that our embedding network has the capability to progressively generate a more discriminative representation in few-shot learning. Moreover, experiments on the task of image classification also show the effectiveness of the proposed design.