Bayesian optimization: Difference between revisions

Revision as of 11:58, 23 September 2024

Revision as of 06:17, 9 October 2024

Line 1:

{{Short description|Statistical optimization technique}}

'''Bayesian optimization''' is a [[sequential analysis|sequential design]] strategy for [[global optimization]] of [[Black box|black-box]] functions,<ref name="Mockus1989"/><ref name="garnett_bayesopt_2022">{{Cite book |last=Garnett |first=Roman |url=https://bayesoptbook.com |title=Bayesian Optimization |date=2023 |publisher=Cambridge University Press |isbn=978-1-108-42578-0 }}</ref><ref name="HennigOsborneKersting2022">{{cite book | author1 = Hennig, P.

| author2 = Osborne, M. A.

| author3 = Kersting, H. P.

Line 9:

| isbn=978-1107163447

| url = https://www.probabilistic-numerics.org/assets/ProbabilisticNumerics.pdf

}}</ref>, that does not assume any functional forms. It is usually employed to optimize expensive-to-evaluate functions. With the rise of [[artificial intelligence]] innovation in the 21st century, Bayesian optimizations have found prominent use in [[machine learning]] problems, for optimizing hyperparameter values.<ref>{{cite journal |first=Jasper |last=Snoek |title=Practical Bayesian Optimization of Machine Learning Algorithms |journal=Advances in Neural Information Processing Systems 25 (NIPS 2012) |year=2012 |url=https://proceedings.neurips.cc/paper/2012/hash/05311655a15b75fab86956663e1819cd-Abstract.html}}</ref><ref>{{cite journal |first=Aaron |last=Klein |title=Fast bayesian optimization of machine learning hyperparameters on large datasets |journal=Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, PMLR |year=2017 |pages=528-536 |url=https://proceedings.mlr.press/v54/klein17a.html}}</ref>

==History==