AWS DeepRacer is a great way to get started with autonomous vehicles using Machine Learning and Reinforcement Learning. While it offers 10 free training hours, costs can rise quickly—especially for those aiming for the Professional League.

In this session, I’ll show you how to use DeepRacer For Cloud (DRFC), a community-built toolset that lets you train models much more affordably using AWS SPOT instances, including those with GPUs and multiple cores. DRFC is based on Docker containers for training, log analysis, track simulation, and more.

We’ll also cover setup tips and how to build a complete, cost-effective training environment so you can compete on equal footing in this unique blend of AI, simulation, and automation.

Train DeepRacer models cheaply with DRFC and SPOT instances. Compete globally in this AI racing challenge.