# What is Bayesian optimization? Definition and meaning

Builds a probabilistic surrogate of an objective and chooses the next point by balancing exploration and exploitation.

## Definition

**Bayesian optimization**: Builds a probabilistic surrogate of an objective and chooses the next point by balancing exploration and exploitation.

In short: Optimizing expensive functions with a surrogate model.

## When it is used

Hyperparameter search and tuning processes where each trial is costly.

## Learn more

The full guide to **[Bayesian optimization](/tools/bayesian-optimization)** explains how it works, when to use it and how it compares with the alternatives, with examples. The [glossary](/glossary) lists every term on the site from A to Z.

Previous term: [Bayesian deep learning](/glossary/bayesian-deep-learning) · Next term: [BentoML](/glossary/bentoml)
