Stanford CS224R Deep Reinforcement Learning | Spring 2025 | Lecture 4: Actor-Critic Methods

Name: Stanford CS224R Deep Reinforcement Learning | Spring 2025 | Lecture 4: Actor-Critic Methods
Uploaded: 2025-12-08T00:00:00.000Z
Duration: 1 h 3 min 30 s
Description: Actor-critic methods improve vanilla policy gradients by fitting a value function (the critic) so the actor can update on advantages—how much better an action is than following the current policy—rather than on a sing...

Stanford Online · 63:30

Actor-critic methods improve vanilla policy gradients by fitting a value function (the critic) so the actor can update on advantages—how much better an action is than following the current policy—rather than on a sing...

Read the full summary on tuber