Study Plan

The second part of this nanodegree program covers value-based methods in deep reinforcement learning and lasts 3 weeks. You can find all of the coding exercises from the lessons in this GitHub repository.

## Weeks 1-2

For the first two weeks, you will learn more about how deep learning is used to increase the complexity of the problems we can solve with reinforcement learning.

Lesson: Deep Q-Networks

In this lesson, you will learn all of the details behind the Deep Q-Networks (DQN) algorithm.

Lesson: Deep Learning with PyTorch (Optional)

We will use PyTorch throughout this program. If the PyTorch framework is new to you, please take a look at our introductory lesson in the extracurricular content.

Lesson: Neural Networks (Optional)

For this course, you are expected to know how neural networks train through backpropagation. If you’d like to review this material, we have prepared this lesson in the extracurricular content.

Lesson: Convolutional Neural Networks (Optional)

If you'd like to review convolutional neural networks, check out this lesson in the extracurricular content.

Lesson: Deep RL in Robotics (Optional)

Learn about how to use the Deep Q-Learning algorithm with real-world robotics from experts at NVIDIA's Deep Learning Institute.

Readings

• Read this scientific article that describes Deep Q-Networks.
• Read the research paper that first introduced the Deep Q-Learning algorithm.

Resources

• Learn more about Deep Q-Learning and Google DeepMind by watching this video.

## Week 3

For the final week, you will focus on the project.