# SAC

An off-policy actor-critic algorithm that maximizes reward plus policy entropy, which encourages exploration and stability.

## What it is used for

Continuous control with good sample efficiency, such as energy and process optimization.