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.
| Original language | English |
|---|---|
| Title of host publication | Automatic Machine Learning |
| Subtitle of host publication | Methods, Systems, Challenges |
| Editors | Frank Hutter, Lars Kotthoff, Joaquin Vanschoren |
| Place of Publication | Cham |
| Publisher | Springer |
| Chapter | 2 |
| Pages | 39-61 |
| Number of pages | 27 |
| ISBN (Electronic) | 978-3-030-05318-5 |
| ISBN (Print) | 978-3-030-05317-8 |
| DOIs | |
| Publication status | Published - 2019 |
Keywords
- meta-learning
- Automatic Machine Learning
Fingerprint Dive into the research topics of 'Meta-learning'. Together they form a unique fingerprint.
Cite this
Vanschoren, J. (2019). Meta-learning. In F. Hutter, L. Kotthoff, & J. Vanschoren (Eds.), Automatic Machine Learning: Methods, Systems, Challenges (pp. 39-61). Springer. https://doi.org/10.1007/978-3-030-05318-5_2