Abstract:
A new model for describing agents' awareness is proposed that generalizes the known reflexion models in the following sense: the proposed model of generalized multiple awareness structures (GMAS) allows for different numbers of participants in agents' perceptions of the situation. Such structures are used to construct information control and can be used to control the dynamics of opinions in social networks and to improve safety in transportation systems even when agents do not know the total number of agents in the system, which is very common in real life but rarely considered in game theory and epistemic logic. In the article, a formal description of probabilistic and interval GMAS is given, and cases of their transformation as a result of observing the choice of actions by agents and as a result of receiving a message by them are considered. Formal definitions of information equilibrium for both types of GMAS are proposed, and the information equilibrium for probabilistic GMAS is formulated in such a way that it allows us to find Bayes –Nash equilibria.
Keywords:decision-making under uncertainty, complex mutual awareness, reflexion, information equilibrium, information processes.
Fulltext:
PDF file (718 kB)