A Unified Framework with Meta-dropout for Few-shot Learning

TL;DR

This paper introduces a unified meta-learning framework with a novel meta-dropout strategy to enhance generalization in few-shot learning tasks across classification and detection datasets.

Contribution

It unifies episodic meta-learning and pre-train finetune methods and proposes meta-dropout to prevent overfitting, improving few-shot learning performance.

Findings

Effective in few-shot object detection and classification

Improves generalization by preventing neural unit co-adaptation

Validated on Pascal VOC, MS COCO, CUB, and mini-ImageNet datasets

Abstract

Conventional training of deep neural networks usually requires a substantial amount of data with expensive human annotations. In this paper, we utilize the idea of meta-learning to explain two very different streams of few-shot learning, i.e., the episodic meta-learning-based and pre-train finetune-based few-shot learning, and form a unified meta-learning framework. In order to improve the generalization power of our framework, we propose a simple yet effective strategy named meta-dropout, which is applied to the transferable knowledge generalized from base categories to novel categories. The proposed strategy can effectively prevent neural units from co-adapting excessively in the meta-training stage. Extensive experiments on the few-shot object detection and few-shot image classification datasets, i.e., Pascal VOC, MS COCO, CUB, and mini-ImageNet, validate the effectiveness of our method.