Reasoning About Belief based on Common Knowledge of Observability of Actions

摘要

Intelligent agents should regard other agents' thoughts. If they can guess what you think with-out asking you, they should do so. We human beings notice other people's thoughts without di-rect communication. One's thoughts depend on one's belief. However, reasoning about belief is very difficult because various factors affect be-lief and they often lead to inconsistency. This paper presents a simple algorithm to calculate multiagent nested belief from an action sequence. The following three factors are essential in this algorithm: 1) the observability conditions of flu-ents and actions, 2) the direct/indirect effects of each action, and 3) the incompatibility of flu-ents, The algorithm performs regressive reason-ing from a query. It has been tested by dozens of examples through a graphic interface. Exper-iments show the system gives plausible answers to various queries. This method will be useful in the construction of plan recognition systems and other advanced systems.