A Voting-Based System for Ethical Decision Making
This addresses ethical decision-making for autonomous systems, but it is incremental as it builds on existing machine learning and social choice methods.
The paper tackles the problem of automating ethical decisions by learning societal preferences and aggregating them to identify desirable choices, achieving a concrete implementation evaluated with preference data from 1.3 million people in the autonomous vehicle domain.
We present a general approach to automating ethical decisions, drawing on machine learning and computational social choice. In a nutshell, we propose to learn a model of societal preferences, and, when faced with a specific ethical dilemma at runtime, efficiently aggregate those preferences to identify a desirable choice. We provide a concrete algorithm that instantiates our approach; some of its crucial steps are informed by a new theory of swap-dominance efficient voting rules. Finally, we implement and evaluate a system for ethical decision making in the autonomous vehicle domain, using preference data collected from 1.3 million people through the Moral Machine website.