Multiagent Reinforcement Learning in Stochastic Games

This repository contains the implementation and visualization of a Reinforcement Learning Framework described in the paper "Multiagent Reinforcement Learning in Stochastic Games" by Junling Hu and Michael P. Wellman for the Paris-Sud FIIL M2 program.

How to run

To build the docker image please run the following bash script:

$ > ./scripts/build.sh

Once the image is built run it with

$ > ./scripts/run.sh

A presentation covering the background, motivation and proof overview for the paper can be found here.