AI engineering glossary · Reinforcement learning
What is actor-critic?
Actor-critic: A family of methods with two parts: an actor that chooses actions, and a critic that estimates values and tells the actor how good its choices were. The critic's feedback lowers variance compared with using raw returns.
Actor-critic: an example
The actor network outputs move probabilities; the critic network outputs V(s); the TD error trains both.
Related terms
- Advantage function
- A2C
- Soft actor-critic
- TD error
More Reinforcement learning terms
- Agent
- AlphaGo
- AlphaZero
- Baseline (policy gradient)
- Behavioural cloning
- Belief state
- Bellman backup
- Bellman equation
Where you learn it
Actor-critic is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.