Information plays a key role in many decision-making situations. This is especially true for cases where there is some uncertainty regarding the different consequences of the decision to be made. In my research, I study the role of information in the context of multiagent systems (MAS) which are very common in real life, e.g., auctions, customers service and even simple human interactions such as selling a car to a potential buyer.
Generally, MAS can be divided into two kinds, cooperative ones, and competitive ones. In competitive MAS information can be used in order to Increase expected profit, gain control or even Shift the system to a more beneficial direction. In cooperative ones, however, information is actually a way to learn about the different participants' strategies, capabilities, intentions, and current state or location. During the talk, I will survey a number of studies dealing both in cooperative and competitive MAS, presenting several novel ways in which information can be used in order to improve different participants' expected profit as well as the social welfare.
