On Generalized Bellman 's Functional Equation.
We show that a recent existence result for the Nash equilibria of generalized games with strategy sets in -spaces is false. We prove that it becomes true if we assume, in addition, that the feasible set of the game (the set of all feasible multistrategies) is closed.
We give a formalization of the ?knowledge games? which allows to study their decidability and convergence as a problem of mathematics. Our approach is based on a metalemma analogous to those of Von Neumann and Morgenstern at the beginning of Game Theory. We are led to definitions which characterize the knowledge games as objects is standard set theory. We then study rigorously the most classical knowledge games and, although we also prove that the ?common knowledge? in these games may be incomputable,...
2000 Mathematics Subject Classification: 91A46, 91A35.We consider nonadaptive search problem for an unknown element x from the set A = {1, 2, 3, . . . , 2^n}, n ≥ 3. For fixed integer S the questions are of the form: Does x belong to a subset B of A, where the sum of the elements of B is equal to S? We wish to find all integers S for which nonadaptive search with n questions finds x. We continue our investigation from [4] and solve the last remaining case n = 2^k , k ≥ 2.
Noncooperative games with systems governed by nonlinear differential equations remain, in general, nonconvex even if continuously extended (i. e. relaxed) in terms of Young measures. However, if the individual payoff functionals are “enough” uniformly convex and the controlled system is only “slightly” nonlinear, then the relaxed game enjoys a globally convex structure, which guarantees existence of its Nash equilibria as well as existence of approximate Nash equilibria (in a suitable sense) for...
We present a negative answer to problem 3.7(b) posed on page 193 of [2], where, in fact, A. Rosłanowski asked: Does every set of Lebesgue measure zero belong to some Mycielski ideal?
Answering a question of Telgársky in the negative, it is shown that there is a space which is β-favorable in the strong Choquet game, but all of its nonempty -subspaces are of the second category in themselves.