Meta-Learning for Multi-Label Few-Shot Classification

TL;DR

This paper addresses the challenge of multi-label few-shot classification by proposing a new benchmark, extending existing solutions, and introducing a neural label count module to improve prediction accuracy on complex datasets.

Contribution

It introduces a multi-label meta-learning benchmark, extends single-label solutions to multi-label scenarios, and proposes a neural module for label count estimation.

Findings

The label propagation algorithm improves multi-label classification accuracy.

The neural label count module enhances the prediction of label numbers.

Combined methods outperform existing approaches on MS-COCO, iMaterialist, and Open MIC datasets.

Abstract

Even with the luxury of having abundant data, multi-label classification is widely known to be a challenging task to address. This work targets the problem of multi-label meta-learning, where a model learns to predict multiple labels within a query (e.g., an image) by just observing a few supporting examples. In doing so, we first propose a benchmark for Few-Shot Learning (FSL) with multiple labels per sample. Next, we discuss and extend several solutions specifically designed to address the conventional and single-label FSL, to work in the multi-label regime. Lastly, we introduce a neural module to estimate the label count of a given sample by exploiting the relational inference. We will show empirically the benefit of the label count module, the label propagation algorithm, and the extensions of conventional FSL methods on three challenging datasets, namely MS-COCO, iMaterialist, and Open MIC. Overall, our thorough experiments suggest that the proposed label-propagation algorithm in conjunction with the neural label count module (NLC) shall be considered as the method of choice.