A Survey on Metric Learning for Feature Vectors and Structured Data

TL;DR

This survey comprehensively reviews metric learning methods for feature vectors and structured data, discussing their advantages, limitations, recent trends, and future challenges in the field.

Contribution

It provides a systematic overview of metric learning approaches, including Mahalanobis, nonlinear, similarity, and local methods, highlighting recent advancements and open challenges.

Findings

Mahalanobis distance is a well-studied framework.

Recent methods include semi-supervised and histogram data metric learning.

Remaining challenges involve structured data and generalization guarantees.

Abstract

The need for appropriate ways to measure the distance or similarity between data is ubiquitous in machine learning, pattern recognition and data mining, but handcrafting such good metrics for specific problems is generally difficult. This has led to the emergence of metric learning, which aims at automatically learning a metric from data and has attracted a lot of interest in machine learning and related fields for the past ten years. This survey paper proposes a systematic review of the metric learning literature, highlighting the pros and cons of each approach. We pay particular attention to Mahalanobis distance metric learning, a well-studied and successful framework, but additionally present a wide range of methods that have recently emerged as powerful alternatives, including nonlinear metric learning, similarity learning and local metric learning. Recent trends and extensions, such as semi-supervised metric learning, metric learning for histogram data and the derivation of generalization guarantees, are also covered. Finally, this survey addresses metric learning for structured data, in particular edit distance learning, and attempts to give an overview of the remaining challenges in metric learning for the years to come.