AJAR: An Argumentation-based Judging Agents Framework for Ethical Reinforcement Learning
Source code for the AJAR framework.
It contains both an implementation of AJAR (argumentation-based judgments), and an environment on which to evaluate AJAR, with learning algorithms.
Citation:
Alcaraz, Benoît, Boissier, Olivier, Chaput, Rémy, & Leturc, Christopher. (2023). AJAR: An Argumentation-based Judging Agents Framework for Ethical Reinforcement Learning (Version v1). The 22nd International Conference on Autonomous Agents and Multiagent Systems (AAMAS' 23), London. Zenodo. https://doi.org/10.5281/zenodo.7628903
BibTeX:
@software{alcaraz_benoit_2023_7628903,
author = {Alcaraz, Benoît and
Boissier, Olivier and
Chaput, Rémy and
Leturc, Christopher},
title = {{AJAR: An Argumentation-based Judging Agents
Framework for Ethical Reinforcement Learning}},
month = mar,
year = 2023,
note = {{The data included in this simulator is a property
of and licensed by OpenEI
(https://doi.org/10.25984/1788456, CC-BY-4.0).
This work is partially funded by the French Région
Auvergne Rhônes-Alpes (AURA), as part of the
Ethics.AI project (Pack Ambition Recherche), and
by the Luxembourg National Research Fund under
Grant No. IS/14717072. We gratefully acknowledge
support from the CNRS/IN2P3 Computing Center (Lyon
— France) for providing computing and data-
processing resources.}},
publisher = {Zenodo},
version = {v1},
doi = {10.5281/zenodo.7628903},
url = {https://doi.org/10.5281/zenodo.7628903}
}