We present preliminary results from our sixth placed entry to the Flatland international competition for train rescheduling, including two improvements for optimized reinforcement learning (RL) training efficiency, and two hypotheses with respect to the prospect of deep RL for complex real-world control tasks: first, that current state of the art policy gradient methods seem inappropriate in the domain of high-consequence environments; second, that learning explicit communication actions (an emerging machine-to-machine language, so to speak) might offer a remedy. These hypotheses need to be confirmed by future work. If confirmed, they hold promises with respect to optimizing highly efficient logistics ecosystems like the Swiss Federal Railways railway network.
@article{arxiv.2004.13439,
title = {Improving Sample Efficiency and Multi-Agent Communication in RL-based Train Rescheduling},
author = {Dano Roost and Ralph Meier and Stephan Huschauer and Erik Nygren and Adrian Egli and Andreas Weiler and Thilo Stadelmann},
journal= {arXiv preprint arXiv:2004.13439},
year = {2020}
}
Comments
Accepted for publication at the 7th Swiss Conference on Data Science (SDS 2020)