Trust region policy optimization
Also known as TRPO. This is the canonical page; those names redirect here.
Pairings in the atlas
- KL trust regionspecialistDeep reinforcement learningmachine-learning
Rivals: other methods for the same problems
- A2C
- A3C
- Actor-critic
- DDPG
- Deep Q-network
- Double DQN
- Double Q-learning
- Dueling DQN
- Expected SARSA
- Generalized advantage estimation
- Hindsight experience replay
- IMPALA
- Monte Carlo control
- PPG
- Policy gradient
- Prioritized experience replay
- Proximal policy optimization
- Q-learning
- Rainbow
- Reward shaping
- SARSA
- Soft actor-critic
- TD3
- Temporal difference learning