Model-Based RL

AI and Machine Learning · World Models · model-based-rl.yaml

Learns the environment's dynamics and plans against them, rather than caching what actions were worth. Buys sample efficiency and pays for it with whatever the model gets wrong.

• supersedescorrects · extends
• classifiesspecializes · part-of
• substitutes forapproximates · alternative-to
• depends onrequires · validates

Colour is the family; a dashed line is the second member of it.

This node

• alternative-tois a competing approach to Q-Learningplans with learned dynamics instead of caching values

Referenced by

• specializesWorld Model is a specific case of thisthe learned model is a latent-space simulator

References

• Dream to Control: Learning Behaviors by Latent Imagination — Hafner, Lillicrap, Ba, Norouzi — 2019 · link