Chapter 9: Deploying Deep RL Agents on Multiple Platforms

This chapter provides recipes to deploy your Deep RL agent models in applications targeting desktop, web, mobile, and beyond. The recipes serve as customizable templates that you can utilize to build and deploy your own Deep RL applications for your use cases. You will also learn how to export RL agent models for serving/deployment in various production-ready formats, such as TensorFlow Lite, TensorFlow.js, and ONNX, and learn how to leverage Nvidia Triton to launch production-ready RL-based AI services.

Specifically, the following recipes are covered in this chapter:

• Packaging Deep RL agents for mobile and IoT devices using TensorFlow Lite
• Deploying RL agents on mobile devices
• Packaging Deep RL agents for the web and Node.js using TensorFlow.js
• Deploying a Deep RL agent as a service
• Packaging Deep RL agents for cross-platform deployments