Introduction to Policy Gradient methods. Learn how to optimize the policy directly without needing a value function, enabling continuous action space control.