arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1908.02488cs.LO
收藏

策略逻辑中的不完全信息处理

Dealing with imperfect information in Strategy Logic

Sophia Knight, Bastien Maubert

摘要:

我们提出了一个扩展的战略逻辑(SL),其中既可以原因不完全信息下的战略和球员的知识。我们的方法的一个原始方面是,我们不强迫策略是统一的,即在语义层面上与参与者的信息一致;相反,我们可以在逻辑本身中表达策略应该是统一的。要做到这一点,我们首先开发了一个“分支时间”版本的SL与完美的信息,我们称之为BSL,其中一个可以量化的不同结果定义的部分分配的策略的球员,这与SL,时间运营商只允许当所有的策略是固定的,只留下一个可能的发挥。接下来,我们进一步扩展BSL添加分布式知识运营商,其语义依赖于部分播放的等价关系。我们得到的逻辑包含了大多数不完全信息的策略逻辑,无论是否是认知的。

英文摘要:

We propose an extension of Strategy Logic (SL), in which one can both reason about strategizing under imperfect information and about players' knowledge. One original aspect of our approach is that we do not force strategies to be uniform, i.e. consistent with the players' information, at the semantic level; instead, one can express in the logic itself that a strategy should be uniform. To do so, we first develop a "branching-time" version of SL with perfect information, that we call BSL, in which one can quantify over the different outcomes defined by a partial assignment of strategies to the players; this contrasts with SL, where temporal operators are allowed only when all strategies are fixed, leaving only one possible play. Next, we further extend BSL by adding distributed knowledge operators, the semantics of which rely on equivalence relations on partial plays. The logic we obtain subsumes most strategic logics with imperfect information, epistemic or not.

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。

↑