A (TU) cooperative game is extendable if every core allocation of each subgame can be extended to a core allocation of the game. It is strongly extendable if any minimal vector in the upper core of any of its subgames can be extended to a core allocation. We prove that strong extendability is...

We introduce the concept of -consistent equilibrium where each player plays a -best response after every history reached with positive probability. In particular, an -consistent equilibrium induces an -equilibrium in any subgame reached along the play path. The existence of -consistent...

The notion of mediated talk in which players communicate through a mediator before starting the game is introduced. It shows that a deterministic mechanized mediator receiving private inputs and producing public outputs can generate any rational correlated equilibrium with rational probabilities...

This note studies relations between lim "V [subscript n]" and the lower long-run average value ("V" [underlined]) in dynamic programming. It is shown that a certain bounded variation conditions of {"V [subscript n]} implies that lim "V [subscript n] [is equal to] "V" [underlined].