AWS DeepRacer: Building a Reinforcement Learning Model
37mIntermediate2022-09-16
Authors

Mike Smith
R&D architect
Course details
Have you ever wanted to find out more about machine learning, but you didn’t know where to begin? AWS DeepRacer is a perfect way to get a fun and useful introduction to machine learning tools, even if you don’t have a technical background. In this course, instructor Michael Smith walks you through the basics of how to use reinforcement learning with AWS DeepRacer—an autonomous one-eighteenth-scale race car designed to test machine learning models by racing on a physical track.
Learn the most important tools for creating your own reinforcement learning algorithm, using built-in DeepRacer resources as your guide. From starting the engine to calibrating the race car and taking the lead on the final lap of the race, Michael helps you understand what reinforcement learning actually does so you can harness it for success in the real world.
Learn the most important tools for creating your own reinforcement learning algorithm, using built-in DeepRacer resources as your guide. From starting the engine to calibrating the race car and taking the lead on the final lap of the race, Michael helps you understand what reinforcement learning actually does so you can harness it for success in the real world.
Skills covered
Machine LearningAmazon Web Services (AWS)AmazonCloud ServicesCloud PlatformsArtificial Intelligence (AI)Cloud ComputingDeep Dive (X:Y)
Concepts
0. Introduction
- 01 - The fast and the curious - Machine learning powered racing
- 02 - What you should know
1. Start Your Engines
- 03 - What is AWS DeepRacer
- 04 - What is reinforcement learning
- 05 - Under the hood of AWS DeepRacer
2. The Fast Straight
- 06 - Assembling AWS DeepRacer
- 07 - Calibrating the car
- 08 - Where the rubber meets the road - The first test drive
- 09 - Building a real-world track
3. Taking the Lead
- 10 - The AWS DeepRacer console
- 11 - Building your first model
- 12 - Training and evaluating your first model
4. The Final Lap
- 13 - The reward function
- 14 - Hyperparameter tuning
- 15 - Incremental training
- 16 - Moving to the real world
Conclusion
- 17 - Next steps - The AWS DeepRacer League