Intrinsic Motivation in HRL

Curiosity-Driven Learning

Learning approach where the agent is motivated by its curiosity, measuring the prediction error of its world model to generate intrinsic exploration rewards.

← Zurück