The Ethics.AI team is a multi-disciplinary and multi-university consortium, composed of permanent researchers, PhD students and interns, experts in AI (machine learning, reinforcement learning, constructivism) and philosophy of science.
The industrial and economic development of Artificial Intelligence (AI) technologies is essential because it allows us to respond to environmental and societal issues of great importance, such as sustainable energy (Smart Grid) or personal assistance, issues that we are targeting in this project, as a…
Continue reading...The EthicsAI project is composed of multiple academic and industrial partners. The research partners at Université Claude Bernard Lyon 1 and the LIRIS lab have expertise in Multi-Agent Systems, Machine Learning (especially Reinforcement Learning), and Cognitive Systems (Self-*). The research partners at École des Mines…
Continue reading...Open-Source software for learning "ethical behaviours". This Python package is a Gym environment that implements a (simple) Smart Grid simulator. The goal is to offer an environment that can be used to experiment and evaluate ethical decision-making algorithms, i.e., algorithms that include ethical considerations, especially…
Continue reading...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.
Continue reading...Publication to a national conference (JFSMA'21).
Available online on HAL.
This (French) paper proposes to combine symbolic judging agents and learning agents in a multi-agent system to learn "ethical behaviors", i.e., behaviors aligned with ethical considerations.
Continue reading...