Using state abstractions to compute personalized contrastive explanations for AI agent behavior
Using state abstractions to compute personalized contrastive explanations for AI agent behavior
复制标题
使用状态抽象来计算人工智能代理行为的个性化对比解释
DOI:
10.1016/j.artint.2021.103570
复制
发表时间:
2021
影响因子:
14.4
通讯作者:
Kambhampati, Subbarao
中科院分区:
文献类型:
--
作者:
Sreedharan, Sarath;Srivastava, Siddharth;Kambhampati, Subbarao
There is a growing interest within the AI research community in developing autonomous systems capable of explaining their behavior to users. However, the problem of computing explanations for users of different levels of expertise has received little research attention. We propose an approach for addressing this problem by representing the user's understanding of the task as an abstraction of the domain model that the planner uses. We present algorithms for generating minimal explanations in cases where this abstract human model is not known. We reduce the problem of generating an explanation to a search over the space of abstract models and show that while the complete problem is NP-hard, a greedy algorithm can provide good approximations of the optimal solution. We empirically show that our approach can efficiently compute explanations for a variety of problems and also perform user studies to test the utility of state abstractions in explanations.
登录
查看更多内容
DOI:
--
发表时间:
2019
期刊:
International Joint Conference on Artificial Intelligence
影响因子:
--
作者:
Sreedharan, Sarath;Srivastava, Siddharth;Smith, David;Kambhampati, Subbarao.
通讯作者:
Kambhampati, Subbarao.
DOI:
--
发表时间:
2019
期刊:
International Joint Conference on Artificial Intelligence
影响因子:
--
作者:
R. Byrne
通讯作者:
R. Byrne
DOI:
--
发表时间:
2003
期刊:
影响因子:
--
作者:
Christopher R. Hitchcock;J. Woodward
通讯作者:
J. Woodward
DOI:
--
发表时间:
--
期刊:
2012 7th ACM/IEEE International Conference on Human-Robot Interaction (HRI)
影响因子:
--
作者:
通讯作者:
--
影响因子:
0.7
作者:
James Webb
通讯作者:
James Webb