Ray Summit

Tutorials

Overview

Wednesday, May 27 is tutorial day. Four, half-day tutorials will be available. There will be two tutorial sessions offered in the morning and two tutorial sessions offered in the afternoon. 

Registration for tutorial day is required to attend the tutorials. To help us plan for the number of attendees for each session, please select the tutorials you plan to attend by adding it to your registration. However, you can switch to the other one in each time slot, if you choose, space permitting.

Morning Tutorials | 9:00 AM – 12:30 PM

Distributed Python with Ray: Hands-on with the core API

Audience for this tutorial 
Attend this tutorial if you are new to Ray, you want to learn about Ray’s capabilities, and you want to explore using Ray to scale your Python applications or Python-based machine learning systems.

Description
Ray is an open-source, distributed framework from U.C. Berkeley’s RISELab that easily scales Python applications from a laptop to a cluster, with an emphasis on the unique performance challenges of ML/AI systems. It is now used in many production deployments.

We’ll use several hands-on examples to explore the problems that Ray solves and the useful features it provides, such as rapid distribution, scheduling, and execution of “tasks” and management of distributed stateful “serverless” computing. We’ll see how it’s used in several ML libraries (and play with examples using those libraries). You’ll learn when to use Ray and how to use it in your projects.

Hands on With Ray Libraries: RLlib, Tune, Serve, and More

Audience for this tutorial 
Attend this tutorial if you are already familiar with Ray, its capabilities and the problems it solves, but you would like a good introduction to the popular ML libraries implemented with Ray: RLlib, Tune, Server, and more as time permits.

Description
This tutorial is for you if you have some familiarity with Ray and how it supports distributed, scalable Python applications will minimal effort, and now you want to learn more about the ML/AI libraries that have been written using Ray: RLlib, Tune, Serve, and more.

Using hands-on examples and exercises, we’ll explore the problems and use cases targeted by each library, the basics of each API, and how you might use it in your applications.

Afternoon Tutorials | 1:30 – 5:00 PM

Hands on Reinforcement Learning with Ray and RLlib

Audience for this tutorial
You are familiar with Ray, for example you attended the morning tutorial, Distributed Python with Ray; Hands-on with the core API. You are interested in reinforcement learning and exploring in some depth how Ray’s RLlib supports it.

Description
Reinforcement learning requires a variety of computational patterns: data processing, simulations, model training, model serving. etc. Few frameworks efficiently support all these patterns at scale.

In this hands-on tutorial, we’ll see how Ray and RLlib seamlessly and efficiently support these workloads, providing an ideal platform for building RL applications. We’ll deep dive into Ray and RLlib APIs. We’ll train and serve an RL-based application.

Third Party Tutorial To Be Announced