Enforcing ethical goals over reinforcement-learning policies

Autor: Emery A. Neufeld, Ezio Bartocci, Agata Ciabattoni, Guido Governatori
Rok vydání: 2022
Předmět:
Zdroj: Ethics and Information Technology. 24
ISSN: 1572-8439
1388-1957
DOI: 10.1007/s10676-022-09665-8
Popis: 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.
Databáze: OpenAIRE
Externí odkaz: