Ein numerisches Verfahren zur Bestimmung maxeminoptimaler Aktionen

For decision problems with only incomplete informations about the distribution of the states the numerical computing of all maxEmin-optimal solutions often is difficult or impossible. In this paper we present a numerical solution method for the computing of all maxEmin-optimal solutions for decision problems with a finite number of actions and states. Having a linear partial information essentially only linear programs must be solved for finding all optimal solutions.