REACT: Revealing Evolutionary Action Consequence Trajectories for Interpretable Reinforcement Learning

Philipp Altmann, Céline Davignon, Maximilian Zorn, Fabian Ritz, Claudia Linnhoff-Popien, Thomas Gabor

2024

Abstract

To enhance the interpretability of Reinforcement Learning (RL), we propose Revealing Evolutionary Action Consequence Trajectories (REACT). In contrast to the prevalent practice of validating RL models based on their optimal behavior learned during training, we posit that considering a range of edge-case trajectories provides a more comprehensive understanding of their inherent behavior. To induce such scenarios, we introduce a disturbance to the initial state, optimizing it through an evolutionary algorithm to generate a diverse population of demonstrations. To evaluate the fitness of trajectories, REACT incorporates a joint fitness function that encourages local and global diversity in the encountered states and chosen actions. Through assessments with policies trained for varying durations in discrete and continuous environments, we demonstrate the descriptive power of REACT. Our results highlight its effectiveness in revealing nuanced aspects of RL models’ behavior beyond optimal performance, with up to 400% increased fidelities, contributing to improved interpretability. Code and videos are available at https://github.com/philippaltmann/REACT.

Paper Citation

in Harvard Style

Altmann P., Davignon C., Zorn M., Ritz F., Linnhoff-Popien C. and Gabor T. (2024). REACT: Revealing Evolutionary Action Consequence Trajectories for Interpretable Reinforcement Learning. In Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: ECTA; ISBN 978-989-758-721-4, SciTePress, pages 127-138. DOI: 10.5220/0013005900003837

in Bibtex Style

@conference{ecta24,
author={Philipp Altmann and Céline Davignon and Maximilian Zorn and Fabian Ritz and Claudia Linnhoff-Popien and Thomas Gabor},
title={REACT: Revealing Evolutionary Action Consequence Trajectories for Interpretable Reinforcement Learning},
booktitle={Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: ECTA},
year={2024},
pages={127-138},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013005900003837},
isbn={978-989-758-721-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: ECTA
TI - REACT: Revealing Evolutionary Action Consequence Trajectories for Interpretable Reinforcement Learning
SN - 978-989-758-721-4
AU - Altmann P.
AU - Davignon C.
AU - Zorn M.
AU - Ritz F.
AU - Linnhoff-Popien C.
AU - Gabor T.
PY - 2024
SP - 127
EP - 138
DO - 10.5220/0013005900003837
PB - SciTePress