Budget-aware Few-shot Learning via Graph Convolutional Network

TL;DR

This paper introduces a budget-aware few-shot learning approach that uses a GCN-based data selection policy to improve data efficiency and classification performance in limited-label scenarios.

Contribution

It proposes a novel meta-learning framework combining a GCN-based data selection policy with few-shot classification, enhancing data efficiency in practical settings.

Findings

Outperforms baseline methods on mini-ImageNet, tiered-ImageNet, and Omniglot datasets.

Demonstrates significant improvements in data efficiency and classification accuracy.

Validates the effectiveness of GCN-based data selection in few-shot learning.

Abstract

This paper tackles the problem of few-shot learning, which aims to learn new visual concepts from a few examples. A common problem setting in few-shot classification assumes random sampling strategy in acquiring data labels, which is inefficient in practical applications. In this work, we introduce a new budget-aware few-shot learning problem that not only aims to learn novel object categories, but also needs to select informative examples to annotate in order to achieve data efficiency. We develop a meta-learning strategy for our budget-aware few-shot learning task, which jointly learns a novel data selection policy based on a Graph Convolutional Network (GCN) and an example-based few-shot classifier. Our selection policy computes a context-sensitive representation for each unlabeled data by graph message passing, which is then used to predict an informativeness score for sequential selection. We validate our method by extensive experiments on the mini-ImageNet, tiered-ImageNet and Omniglot datasets. The results show our few-shot learning strategy outperforms baselines by a sizable margin, which demonstrates the efficacy of our method.