Multi-Label Zero-Shot Learning with Transfer-Aware Label Embedding Projection

TL;DR

This paper introduces a transfer-aware label embedding projection method for multi-label zero-shot learning, effectively transferring knowledge from seen to unseen labels and improving multi-label image classification performance.

Contribution

It proposes a novel label embedding projection technique that enhances inter-label relationships and facilitates knowledge transfer in multi-label zero-shot learning.

Findings

The approach improves zero-shot multi-label classification accuracy.

Auxiliary information enhances label relation structures.

Experimental results demonstrate the method's effectiveness.

Abstract

Zero-shot learning transfers knowledge from seen classes to novel unseen classes to reduce human labor of labelling data for building new classifiers. Much effort on zero-shot learning however has focused on the standard multi-class setting, the more challenging multi-label zero-shot problem has received limited attention. In this paper we propose a transfer-aware embedding projection approach to tackle multi-label zero-shot learning. The approach projects the label embedding vectors into a low-dimensional space to induce better inter-label relationships and explicitly facilitate information transfer from seen labels to unseen labels, while simultaneously learning a max-margin multi-label classifier with the projected label embeddings. Auxiliary information can be conveniently incorporated to guide the label embedding projection to further improve label relation structures for zero-shot knowledge transfer. We conduct experiments for zero-shot multi-label image classification. The results demonstrate the efficacy of the proposed approach.