Inverse Norm Conflict Resolution

Inverse Norm Conflict Resolution
复制标题

逆范数冲突解决

DOI:
10.1145/3278721.3278775
复制
发表时间:
2018
期刊:
Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society
影响因子:
--
通讯作者:
Matthias Scheutz
Matthias Scheutz
中科院分区:
--
文献类型:
--
作者:
Daniel Kasenberg;Matthias Scheutz

文献摘要

参考文献

被引文献

相似文献

在以前的工作中,我们提供了一种“规范冲突解决”算法,允许随机域中的代理(由马尔可夫决策过程表示)“最大限度地满足”一组道德或社会规范,其中这些规范由线性时序逻辑(LTL)中的语句表示。这需要代理设计者提供权重,具体说明每个规范的相对重要性。在本文中,我们提出了一种“逆范数冲突消解”算法,用于从论证中学习这些权重。该方法基于编码观察行为的策略与代表最优范数遵循行为的策略之间的相对熵来最小化代价函数。我们在简单的GridWorld域中证明了该算法的有效性。
In previous work we provided a "norm conflict resolution" algorithm allowing agents in stochastic domains (represented by Markov Decision Processes) to "maximally satisfy" a set of moral or social norms, where such norms are represented by statements in linear temporal logic (LTL). This required the agent designer to provide weights specifying the relative importance of each norm. In this paper, we propose an "inverse norm conflict resolution'' algorithm for learning these weights from demonstration. This approach minimizes a cost function based on the relative entropy between a policy encoding the observed behavior and a policy representing optimal norm-following behavior. We demonstrate the effectiveness of the algorithm in a simple GridWorld domain.
随机域中的范数冲突解决
DOI: --
发表时间: 2018
期刊: Proceedings of the ... AAAI Conference on Artificial Intelligence
影响因子: --
作者:
Kasenberg, Daniel;Scheutz, Matthias
通讯作者: Scheutz, Matthias