Study Plan

The third part of this nanodegree program covers policy-based methods in deep reinforcement learning. You can find all of the coding exercises from the lessons in this GitHub repository.

## Lessons

Lesson: Introduction to Policy-Based Methods

In this lesson, you will learn about methods such as hill climbing, simulated annealing, and adaptive noise scaling. You'll also learn about cross-entropy methods and evolution strategies.

Lesson: Policy Gradient Methods

In this lesson, you'll study REINFORCE, along with improvements we can make to lower the variance of policy gradient algorithms.

Lesson: Proximal Policy Optimization

In this lesson, you'll learn about Proximal Policy Optimization (PPO), a cutting-edge policy gradient method.

Lesson: Actor-Critic Methods

In this lesson, you'll learn how to combine value-based and policy-based methods, bringing together the best of both worlds, to solve challenging reinforcement learning problems.

Lesson: Deep RL for Finance (Optional)

In this optional lesson, you'll learn how to apply deep reinforcement learning techniques for optimal execution of portfolio transactions.

Resources (Optional)

• Read the most famous blog post on policy gradient methods.
• Implement a policy gradient method to win at Pong in this Medium post.
• Learn more about evolution strategies from OpenAI.