Lecture 11: Deep Reinforcement Learning

(Big slides Download Big slides) (Small slides Download Small slides) (Recording)

This lecture is about

• Approximate Value Iteration using Neural nets
• Policy Gradient and Actor-critic methods
• Deep Deterministic Policy Gradient Method
• Example: Verfication of Rocket Controlled by Neural Network

Read sections 5.3, 5.4.1, 5.7.1, 2.4.2 in Bertsekas' book Links to an external site.