A Voting-Based System for Ethical Decision Making
A Voting-Based System for Ethical Decision Making
复制标题
DOI：
10.1609/aaai.v32i1.11512
复制
发表时间：
2017-09
期刊：
影响因子：
--
通讯作者：
Ritesh Noothigattu;Snehalkumar `Neil' Gaikwad;E. Awad;Sohan Dsouza;Iyad Rahwan;Pradeep Ravikumar;Ariel D. Procaccia
中科院分区：
文献类型：
--
作者：
Ritesh Noothigattu;Snehalkumar `Neil' Gaikwad;E. Awad;Sohan Dsouza;Iyad Rahwan;Pradeep Ravikumar;Ariel D. Procaccia
我们提出了一个通用的方法来自动化道德决策，利用机器学习和计算社会选择。简而言之，我们建议学习一个社会偏好模型，并且当在运行时面临特定的道德困境时，有效地聚合这些偏好以确定理想的选择。我们提供了一个具体的算法，实例化我们的方法，它的一些关键步骤是由一个新的理论交换优势有效的投票规则。最后，我们使用通过道德机器网站收集的130万人的偏好数据，在自动驾驶汽车领域实现和评估道德决策系统。
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.