AI engineering glossary · Reinforcement learning
What is DQN (Deep Q-network)?
Deep Q-network (also called DQN): A method that uses a neural network to estimate action values from raw observations such as game pixels, trained with Q-learning targets. It relies on experience replay and a target network for stability, and it showed human-level play on many Atari games.
Deep Q-network: an example
A network takes 4 stacked 84x84 frames and outputs 18 values, one per joystick action; the highest is picked.
Related terms
More Reinforcement learning terms
- Deterministic policy
- Deterministic policy gradient
- Discount factor
- Distributional reinforcement learning
- Double DQN
- Double Q-learning
- Dueling network
- Dyna
Where you learn it
Deep Q-network is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.