Charalambos D. Charalambous - Communication systems
Date: - Location: Eurecom
Abstract: Past dynamic programming (DP) studies on multiagent decentralized stochastic optimal control, with delayed sharing information patterns, focused on a single value function conditioned on the shared or common information of all controls or agents. This approach has several limitations, and does not retain key fundamental properties of centralized DP of partially observable Markov decision problems (POMDPs). In this talk, I will present a new generalized DP framework based on decentralized team equilibrium called Person-by-Person (PbP) optimality in static team theory. Each agent is assigned an individual value function conditioned on the agent's delayed sharing information pattern, while all other agents are fixed to their optimal responses. I will introduce several new DP equations which characterize decentralized team equilibrium, with emphasis on the role of private and common information components of each agent's information pattern to reduce complexity and to retain the key fundamental properties of centralized DP equations of POMDPs: 1) the optimization is over the agent's action spaces rather than their strategy spaces, 2) each agent compresses the data into a private information state, and 2) a centralized information state which is common to all agents. The new DP framework settles a long standing open problem, since H. Witsenhausen’s seminal paper, “Separation of estimation and control for discrete time systems,'' in Proceedings of the IEEE, vol.59, no.11, pp.1557—1566, 1971. Short bio: Charalambos D. Charalambous received his B.S., M.E., and Ph.D. in 1987,1988, and 1992, respectively, all from the Department of Electrical Engineering, Old Dominion University, Virginia, USA. In 2003 he joined the Department of Electrical and Computer Engineering, University of Cyprus. He was an Associate Professor at University of Ottawa, from 1999 to 2003, and served on the faculty of McGill University, Department of Electrical and Computer Engineering, as a non-tenure faculty member, from 1995 to 1999. From 1993 to 1995 he was a post-doctoral fellow at Idaho State University. He is currently editor at large of the Journal Mathematics of Control, Signals and Systems and on the editorial board of Entropy, Information Theory, Probability and Statistics Section. In the past he served as an Associate Editor of the IEEE Transactions on Automatic Control, Systems and Control Letters, and IEEE Communications Letters. Charalambous' research is mainly focused on Stochastic control, estimation and decision, information theory of stochastic systems, optimization of stochastic systems subject to ambiguity, decentralized stochastic games of control with asymmetry of information, and their applications to networked control and communication systems.
