Hierarchical Relationship Alignment Metric Learning

TL;DR

This paper introduces HRAML, a hierarchical deep metric learning model that aligns relationships in feature and label spaces, improving performance in complex multi-label and distribution learning scenarios.

Contribution

It combines deep learning with the RAML framework to model complex relationships, extending linear metrics to hierarchical, non-linear models for diverse tasks.

Findings

HRAML outperforms popular methods and RAML in experiments.

The hierarchical approach effectively models complex datasets.

Relationship alignment improves metric learning accuracy.

Abstract

Most existing metric learning methods focus on learning a similarity or distance measure relying on similar and dissimilar relations between sample pairs. However, pairs of samples cannot be simply identified as similar or dissimilar in many real-world applications, e.g., multi-label learning, label distribution learning. To this end, relation alignment metric learning (RAML) framework is proposed to handle the metric learning problem in those scenarios. But RAML learn a linear metric, which can't model complex datasets. Combining with deep learning and RAML framework, we propose a hierarchical relationship alignment metric leaning model HRAML, which uses the concept of relationship alignment to model metric learning problems under multiple learning tasks, and makes full use of the consistency between the sample pair relationship in the feature space and the sample pair relationship in the label space. Further we organize several experiment divided by learning tasks, and verified the better performance of HRAML against many popular methods and RAML framework.