Learning to Learn in a Semi-Supervised Fashion

TL;DR

This paper introduces a meta-learning approach for semi-supervised learning that transfers semantics-oriented similarity representations from labeled to unlabeled data, improving performance in tasks like re-identification and image retrieval.

Contribution

It proposes a novel meta-learning scheme that leverages labeled data to derive similarity representations and transfer them to unlabeled data, addressing disjoint label sets.

Findings

Outperforms state-of-the-art methods across various tasks.

Effectively transfers semantic similarity representations.

Enhances semi-supervised learning performance.

Abstract

To address semi-supervised learning from both labeled and unlabeled data, we present a novel meta-learning scheme. We particularly consider that labeled and unlabeled data share disjoint ground truth label sets, which can be seen tasks like in person re-identification or image retrieval. Our learning scheme exploits the idea of leveraging information from labeled to unlabeled data. Instead of fitting the associated class-wise similarity scores as most meta-learning algorithms do, we propose to derive semantics-oriented similarity representations from labeled data, and transfer such representation to unlabeled ones. Thus, our strategy can be viewed as a self-supervised learning scheme, which can be applied to fully supervised learning tasks for improved performance. Our experiments on various tasks and settings confirm the effectiveness of our proposed approach and its superiority over the state-of-the-art methods.