AI engineering glossary · Reinforcement learning
What is model-free?
Model-free: Describes methods that learn what to do directly from experience, without building an explicit model of the transition probabilities or scores. They are simpler and robust to model errors, but usually need much more interaction data.
Model-free: an example
Q-learning and REINFORCE never estimate how the world responds; they only use observed transitions.
Related terms
More Reinforcement learning terms
- Monte Carlo methods (reinforcement learning)
- Monte Carlo tree search
- Multi-agent reinforcement learning
- Multi-armed bandit
- MuZero
- n-step return
- Nash equilibrium
- Natural policy gradient
Where you learn it
Model-free is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.