Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges
Abstrak
Most machine learning algorithms are configured by a set of hyperparameters whose values must be carefully chosen and which often considerably impact performance. To avoid a time‐consuming and irreproducible manual process of trial‐and‐error to find well‐performing hyperparameter configurations, various automatic hyperparameter optimization (HPO) methods—for example, based on resampling error estimation for supervised machine learning—can be employed. After introducing HPO from a general perspective, this paper reviews important HPO methods, from simple techniques such as grid or random search to more advanced methods like evolution strategies, Bayesian optimization, Hyperband, and racing. This work gives practical recommendations regarding important choices to be made when conducting HPO, including the HPO algorithms themselves, performance evaluation, how to combine HPO with machine learning pipelines, runtime improvements, and parallelization.
Topik & Kata Kunci
Penulis (12)
B. Bischl
Martin Binder
Michel Lang
Tobias Pielok
J. Richter
Stefan Coors
Janek Thomas
Theresa Ullmann
Marc Becker
A. Boulesteix
Difan Deng
M. Lindauer
Akses Cepat
- Tahun Terbit
- 2021
- Bahasa
- en
- Total Sitasi
- 856×
- Sumber Database
- Semantic Scholar
- DOI
- 10.1002/widm.1484
- Akses
- Open Access ✓