RL-Viz provides the reinforcement learning community for the first time ever with a flexible, general, standardized, cross language and cross platform protocol/framework for managing and visualizing the interaction between agents and environments in reinforcement learning experiments. RL-Viz is a p...

Reinforcement learning toolkit written in Java, with support for multi-agent environments. *Current features* * multi-agent support * standard RL algorithms: TD(lambda), Sarsa(lambda), Q(0), actor-critic(lambda). * configurable grid world * configurable, physically realistic continu...

This project is based on my seminar paper on reinforcement learning. There will be some examples of reinforcement learning algorithms such as QLearning or SARSA.

My reinforcement learning efforts. As for now, just my robot control with reinforcement learning undergraduate thesis.

This project is about using Reinforment Learning, a field of machine learning in a FPS video game.

RL-Glue is an open source machine learning project that makes it easy for the reinforcement learning research community to connect and reuse agents and environments written in a variety of programming languages. For full details, please see the RL-Glue Homepage [http://glue.rl-community.org/]. ...

Code providing an environment for an agent to interact with Pacman and learn how to play it. The agent is currently reimplementing the cross-entropy approach taken by Szita and Lorincz.

This is the collection of projects I maintain under rlglue. It includes development of agents and environments.

Basic statistics and inference for MDPs, algorithms for reinforcement learning.