【70周年校庆系列学术报告】Zero-sum semi-Markov games under the probability criterion

2021年10月13日 10:00-12:00  腾讯会议：994 136 255

This talk concerns with zero-sum semi-Markov games under a discounted probability criterion. With a given goal level, two players are interested in the probability that the discounted reward/cost obtained during infinite horizon is no less than the given level for player 1/2, which is regarded as the security/risk probability for player 1/2. First, we give a characterization of the discounted probability. Then, we establish the associated Shapley equation for the discounted probability criterion, and further show the existence of a Nash equilibrium and an iteration algorithm for computing Nash equilibria. Finally, we illustrate our main results by a business system.