arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1603.04586cs.AIcs.ROcs.SYeess.SY
收藏

通过混合可观测域中的多臂老虎机放松实现最优感知

Optimal Sensing via Multi-armed Bandit Relaxations in Mixed Observability Domains

Tampere University of Technology(坦佩雷理工大学)

Mikko Lauri, Risto Ritala

AI总结:

研究在混合可观测域中不确定决策问题,通过放松约束推导最优价值函数上界,并利用多臂老虎机的可计算最优策略提升搜索空间剪枝效率,实验显示在目标监控领域有效。

摘要:

在混合可观测域中研究不确定决策问题,目标是在部分可观测随机过程中,在完全可观测内部状态约束下最大化获得的信息量。通过放松约束推导最优价值函数的上界,识别出在何种条件下放松问题可转化为多臂老虎机,其最优策略易于计算。将该上界应用于原始问题的搜索空间剪枝,并通过模拟实验评估对解质量的影响。实验结果表明,在目标监控领域有效剪枝了搜索空间。

英文摘要:

Sequential decision making under uncertainty is studied in a mixed observability domain. The goal is to maximize the amount of information obtained on a partially observable stochastic process under constraints imposed by a fully observable internal state. An upper bound for the optimal value function is derived by relaxing constraints. We identify conditions under which the relaxed problem is a multi-armed bandit whose optimal policy is easily computable. The upper bound is applied to prune the search space in the original problem, and the effect on solution quality is assessed via simulation experiments. Empirical results show effective pruning of the search space in a target monitoring domain.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑