Get Rolling on Machine Learning
summary
Hands-on walkthrough of building and training an AWS DeepRacer model using AWS RoboMaker for 3D simulation and Amazon SageMaker for training, with experiences exchanged via S3 and video via Kinesis. Describes the AWS DeepRacer EVO hardware upgrade (stereo camera + LiDAR), the console’s Python reward function authoring, a six-layer CNN + policy network trained with Proximal Policy Optimization, typical convergence around 20 episodes, tunable action space/garage sensors, and how models compete in Open/Pro League prizes up to re:Invent.
AI-written summary · about
chapters (8)
- 0:00Introduction to speaker
- 1:08What is AWS DeepRacer
- 3:21Machine learning basics
- 6:58Reinforcement learning explained
- 12:58Training process in DeepRacer
- 16:35DeepRacer console architecture
- 21:59AWS DeepRacer League 2021
- 23:21Resources and next steps
related talks
23mGet rolling with machine learning on AWS DeepRacer
27mGet rolling with machine learning on AWS DeepRacer
29mGet rolling with ML on AWS DeepRacer
37mAWS DeepRacer: Train, Evaluate, and Tune
37mAWS DeepRacer:train, Evaluate, and Tune Your Reinforcement Learning Model
27mReplicate AWS DeepRacer architecture to conquer the track with SageMaker
42mFrom Zero to Machine Learning in 45 Minutes
29mKickstart Your Machine Learning Career with AWS DeepRacer