An integrated approach to solving influence diagrams and finite-horizon partially observable decision processes
An integrated approach to solving influence diagrams and finite-horizon partially observable decision processes
复制标题
求解影响图和有限范围部分可观察决策过程的集成方法
DOI:
10.1016/j.artint.2020.103431
复制
发表时间:
2021
影响因子:
14.4
通讯作者:
Hansen, Eric A.
中科院分区:
文献类型:
--
作者:
Hansen, Eric A.
We show how to integrate a variable elimination approach to solving influence diagrams with a value iteration approach to solving finite-horizon partially observable Markov decision processes (POMDPs). The integration of these approaches creates a variable elimination algorithm for influence diagrams that has much more relaxed constraints on elimination order, which allows improved scalability in many cases. The new algorithm can also be viewed as a generalization of the value iteration algorithm for POMDPs that solves non-Markovian as well as Markovian problems, in addition to leveraging a factored representation for improved efficiency. The development of a single algorithm that integrates and generalizes both of these classic algorithms, one for influence diagrams and the other for POMDPs, unifies these two approaches to solving Bayesian decision problems in a way that combines their complementary advantages.
登录
查看更多内容
DOI:
--
发表时间:
2000
期刊:
International Conference on Artificial Intelligence Planning Systems
影响因子:
--
作者:
R. Dechter
通讯作者:
R. Dechter
DOI:
--
发表时间:
2017
期刊:
Conference on Uncertainty in Artificial Intelligence
影响因子:
--
作者:
Manuel Luque;M. Arias;F. Díez
通讯作者:
F. Díez
影响因子:
1.7
作者:
M. Arias;F. Díez
通讯作者:
F. Díez
DOI:
--
发表时间:
2020
期刊:
PMLR
影响因子:
--
作者:
Hansen, Eric A.;Bowman, Thomas
通讯作者:
Bowman, Thomas
DOI:
--
发表时间:
1985
期刊:
影响因子:
--
作者:
J. A. Tatman
通讯作者:
J. A. Tatman