Repository files navigation

Thesereinforcement learning tutorials use environments fromOpenAI Gym to illustrate how to train policiesinRLlib.

To get started usegit to clone this public repository:

git clone https://github.com/DerwenAI/rllib_tutorials.gitcd rllib_tutorials

Then usepip to install the required dependencies:

python3 -m pip install -U pippython3 -m pip install -r requirements.txt

Alternatively, if you useconda for installing Python packages:

conda create -n rllib_tutorials python=3.7conda activate rllib_tutorialspython3 -m pip install -r requirements.txt

UseJupyterLab to run thenotebooks.Connect into the directory for this repo, then launch JupyterLab with thecommand line:

jupyter-lab

Intro to Reinforcement Learning and Tour Through RLlib covers anintroductory, hands-on coding tour through RLlib and relatedcomponents of Ray used for reinforcement learning applications inPython.This webinar begins with a lecture that introduces reinforcementlearning, including the essential concepts and terminology, plus showtypical coding patterns used in RLlib.We'll also explore four different well-known reinforcement learningenvironments through hands-on coding.The intention is to compare and contrast across these environments tohighlight the practices used in RLlib.Then we'll follow with Q&A.

• some Python programming experience
• some familiarity with machine learning
• clone/install the Git repo
• no previous work in reinforcement learning
• no previous hands-on experience with RLlib

See also:

• Intro to RLlib: Example Environments

Using Reinforcement Learning begins with a brief tutorial about how tobuild custom Gym environments to use with RLlib, to use as a starting point.We’ll then explore hands-on coding for RL through two use cases:

1. Contextual bandits with a financial portfolio optimization example–a real-world problem addressed with a “constrained” class of RL algorithms
2. Building a recommender system with RLlib–new approaches to recommenders, which can be adapted to similar use cases

• Some Python programming experience
• Some familiarity with machine learning
• Clone/install the Git repo
• Intro to Reinforcement Learning and Tour Through RLlib or equivalent

• Forum
• GitHub
• Twitter
• Slack

Ray Summit
June 22-24, 2021
online, free registration
https://www.anyscale.com/ray-summit-2021