# What is Optimization? Definition and meaning

The study of algorithms such as gradient descent that minimize objectives under constraints.

## Definition

**Optimization**: The study of algorithms such as gradient descent that minimize objectives under constraints.

In short: Finding the parameters that minimize a loss.

## When it is used

Choosing optimizers, learning-rate schedules and understanding training stability.

## Learn more

The full guide to **[Optimization](/tools/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: [OpenTelemetry](/glossary/opentelemetry) · Next term: [Optimizers](/glossary/optimizers)
