Stanford CS224R Deep Reinforcement Learning | Spring 2025 | Lecture 3: Policy Gradients

Name: Stanford CS224R Deep Reinforcement Learning | Spring 2025 | Lecture 3: Policy Gradients
Uploaded: 2025-12-08T00:00:00.000Z
Duration: 1 h 2 min 37 s
Description: Policy gradients (REINFORCE) are the first online RL algorithm in this lecture: after you initialize a policy (randomly, from imitation, or with heuristics), you collect rollouts, estimate \(\nabla\theta J(\theta)\) w...

Stanford Online · 62:37

Policy gradients (REINFORCE) are the first online RL algorithm in this lecture: after you initialize a policy (randomly, from imitation, or with heuristics), you collect rollouts, estimate \(\nabla\theta J(\theta)\) w...

Read the full summary on tuber