Meta-agent programs

There are numerous applications where an agent a needs to reason about the beliefsof another agent, as well as about the actions that other agents may take. In (21)the concept of an agent program is introduced, and a language within which theoperating principles of an agent can be declaratively encoded on top of imperativedata structures is defined. In this paper we first introduce certain belief data struc-tures that an agent needs to maintain. Then we introduce the concept of a MetaAgent Program (map), that extends the framework of (21; 19), so as to allow agentsto perform metareasoning. We build a formal semantics for maps, and show howthis semantics supports not just beliefs agent a may have about agent b’s state, butalso beliefs about agents b’s beliefs about agent c’s actions, beliefs about b’s beliefsabout agent c’s state, and so on. Finally, we provide a translation that takes anymap as input and converts it into an agent program such that there is a one-onecorrespondence between the semantics of the map and the semantics of the resultingagent program. This correspondence allows an implementation of maps to be builton top of an implementation of agent programs.