This repository contains the implemantation of the learning algorithm for https://arxiv.org/abs/2210.05393.
RL agents are heuristics for the On-the-fly DCS algorithm implemented at MTSA.
This is a work in progress at Lafhis, University of Buenos Aires.
- Run the following command to install the required python modules.
python -m pip install -r requirements.txt
- Download the mtsa.jar file from here and move it to the directory of this readme.
The experiments from the paper can be run with the following command, executing it from the repo's root directory.
python src/main.py --exp-path "my experiment"All results are then stored at experiments/results/my experiment.