TRPO — Trust Region Policy Optimization

Fact

An RL algorithm that limits how far a policy can change at once.

In Plain Words

TRPO is bowling with bumpers for a learning robot. It can roll better, but not launch the ball into lane 6.

It makes policy updates safer. You meet it in RL work for robots and game control.

Related Concepts

• Reinforcement Learning
• Policy Gradient
• Proximal Policy Optimization
• Kullback-Leibler Divergence

RL
TRPO is an RL method for updating a policy safely.

Policy Gradient
TRPO adds a safety boundary to Policy Gradient updates.

PPO
PPO is a simpler version inspired by TRPO.

KL Divergence
TRPO uses KL Divergence to keep the new policy close to the old one.