Scenario-assisted Deep Reinforcement Learning

Raz Yerushalmi, Guy Amir, Achiya Elyasaf, David Harel, Guy Katz, Assaf Marron

2022

Abstract

Deep reinforcement learning has proven remarkably useful in training agents from unstructured data. However, the opacity of the produced agents makes it difficult to ensure that they adhere to various requirements posed by human engineers. In this work-in-progress report, we propose a technique for enhancing the reinforcement learning training process (specifically, its reward calculation), in a way that allows human engineers to directly contribute their expert knowledge, making the agent under training more likely to comply with various relevant constraints. Moreover, our proposed approach allows formulating these constraints using advanced model engineering techniques, such as scenario-based modeling. This mix of black-box learning-based tools with classical modeling approaches could produce systems that are effective and efficient, but are also more transparent and maintainable. We evaluated our technique using a case-study from the domain of internet congestion control, obtaining promising results.

Paper Citation

in Harvard Style

Yerushalmi R., Amir G., Elyasaf A., Harel D., Katz G. and Marron A. (2022). Scenario-assisted Deep Reinforcement Learning. In Proceedings of the 10th International Conference on Model-Driven Engineering and Software Development - Volume 1: MODELSWARD, ISBN 978-989-758-550-0, pages 310-319. DOI: 10.5220/0010904700003119

in Bibtex Style

@conference{modelsward22,
author={Raz Yerushalmi and Guy Amir and Achiya Elyasaf and David Harel and Guy Katz and Assaf Marron},
title={Scenario-assisted Deep Reinforcement Learning},
booktitle={Proceedings of the 10th International Conference on Model-Driven Engineering and Software Development - Volume 1: MODELSWARD,},
year={2022},
pages={310-319},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010904700003119},
isbn={978-989-758-550-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 10th International Conference on Model-Driven Engineering and Software Development - Volume 1: MODELSWARD,
TI - Scenario-assisted Deep Reinforcement Learning
SN - 978-989-758-550-0
AU - Yerushalmi R.
AU - Amir G.
AU - Elyasaf A.
AU - Harel D.
AU - Katz G.
AU - Marron A.
PY - 2022
SP - 310
EP - 319
DO - 10.5220/0010904700003119