All Posts

Ray Summit, the annual Ray user conference, is back this August, and the Call for Papers is open until April 18! In this blog post, we’ll give you all the information you need to prepare a stellar talk proposal.

We’re holding the first-ever Production RL Summit on March 29. Read more about our featured speakers and sessions and learn why you should join if you’re a reinforcement learning practitioner — or even if you have just a passing interest in RL.

As of version 1.11, Redis is no longer Ray’s default metadata store or pub/sub message broker. In this post, we’ll cover the history of Redis in Ray and how this change allows us to add better support for fault tolerance and high availability.

Ray 1.11 is here! Ray no longer starts Redis by default, opening up the possibility for better support of fault tolerance and high availability in future releases. Plus, there’s a new, more intuitive design for the Ray docs.

Implementing RL algorithms without understanding the limitations of Deep Q Networks can lead to poor results. In this post, we'll cover two important limitations that can make Q learning unstable as well as practical ways to resolve these issues.

In this article, we’ll cover how to deploy XGBoost with two frameworks: Flask and Ray Serve. We’ll also highlight the advantages of Ray Serve over other serving solutions when comparing models in production.

Next in our reinforcement learning series, we'll derive the Q learning algorithm and show how it was applied to yield one of the first breakthroughs that started the field of Deep RL: the Deep Q Network.

McKinsey’s QuantumBlack helps Team New Zealand retool the way they train with reinforcement learning built on Ray.

Previously, we provided an informal introduction to how reinforcement learning (RL) works and how it’s used in the real world. In this post, we will describe how the RL mathematical framework actually works and derive it from first principles.

In this article, we will highlight the options available for serving a PyTorch model into production and deploying it with several frameworks, such as TorchServe, Flask, and FastAPI.