This repository contains a translator from instances of the Moving Firefighter Problem (MFP) into the Bulldozer Reinforcement Learning environment. The MFP is a problem derived from graph theory. Here, we provide an implementation to translate instances of the MFP into an environment suitable for solving with Deep Reinforcement Learning using the Bulldozer framework.
We share the results of running both algorithms (MFP and Bulldozer) on translated instances: