Neufeld, E. A., Bartocci, E., Ciabattoni, A., & Governatori, G. (2022). Enforcing ethical goals over reinforcement-learning policies. Ethics and Information Technology, 24(4), Article 43. https://doi.org/10.1007/s10676-022-09665-8
Recent years have yielded many discussions on how to endow autonomous agents with the ability to make ethical decisions, and the need for explicit ethical reasoning and transparency is a persistent theme in this literature. We present a modular and transparent approach to equip autonomous agents with the ability to comply with ethical prescriptions, while still enacting pre-learned optimal behaviour. Our approach relies on a normative supervisor module, that integrates a theorem prover for defeasible deontic logic within the control loop of a reinforcement learning agent. The supervisor operates as both an event recorder and an on-the-fly compliance checker w.r.t. an external norm base. We successfully evaluated our approach with several tests using variations of the game Pac-Man, subject to a variety of “ethical” constraints.
en
Project title:
Werkzeuge für logisches Schließen in der Deontischen Logik und Anwendungen auf heilige indische Schriften: MA16-028 (WWTF Wiener Wissenschafts-, Forschu und Technologiefonds)
CC BY 4.0
