Meta-Learning: A Survey

TL;DR

This survey reviews the field of meta-learning, highlighting how it enables faster learning of new tasks by leveraging experience, thereby improving machine learning efficiency and automating algorithm design.

Contribution

It provides a comprehensive overview of current meta-learning methods, emphasizing their potential to accelerate learning and replace manual algorithm engineering.

Findings

Meta-learning improves learning speed across diverse tasks.

It enables automation of algorithm design.

The field is rapidly evolving with new approaches.

Abstract

Meta-learning, or learning to learn, is the science of systematically observing how different machine learning approaches perform on a wide range of learning tasks, and then learning from this experience, or meta-data, to learn new tasks much faster than otherwise possible. Not only does this dramatically speed up and improve the design of machine learning pipelines or neural architectures, it also allows us to replace hand-engineered algorithms with novel approaches learned in a data-driven way. In this chapter, we provide an overview of the state of the art in this fascinating and continuously evolving field.