ABSTRACT
Multi-Objective Multiagent Planning (MOMAP) addresses the problem of resolving conflicts between individual agent interests and the group interests. In this paper, we address this problem by presenting a formal framework to represent objective relationships, a decision model using a Vector-Valued Decentralized Markov Decision Process (2V-DEC-MDP) and an algorithm to solve the resulting 2V-DEC-MDP. The formal framework of a Vector-Valued MDP considered uses the value function which returns a vector representing the individual and the group interests. An optimal policy in such contexts is not clear but in this approach we develop a regret-based technique to find a good tradeoff between the group and individual interests. To do that, the approach we present uses Egalitarian Social Welfare orderings that allow an agent to consider during its local optimization the satisfaction of all criteria and reducing their differences. The obtained result is a good balance between individual and group satisfactions where the local policies can lead to more global satisfying behaviors in some settings. This result is illustrated in many examples and compared to alternate local policies.
- A. Beynier and A. Mouaddib. A polynomial algorithm to solve decentralized mdp with temporal constraints. In AAMAS, 2005. Google ScholarDigital Library
- C. Boutilier, F. Bacchus, and R. Brafman. Ucp-networks: A directed graphical representation of conditional utilities. In International Conference on Uncertainty in Artificial Intelligence (UAI), 2001. Google ScholarDigital Library
- S. Jiaying, Z. Xiaoqin, and V. Lesser. Degree of local coordiantion and its implication on global utility. In Proceedings of AAMAS, pages 546--553, 2004. Google ScholarDigital Library
- R. Keeney and H. Raiffa. Decision with Multiple Objectives: Preferences and Value Tradeoffs. John Wiley and Sons, Inc., 1976.Google Scholar
- A. Mouaddib. Anytime coordination for progressive planning agents. In AAAI-99, pages 564--569, 1999. Google ScholarDigital Library
- P. Perny, O. Spanjaard, and P. Weng. Algebraic markov decision processes. In IJCAI05, 2005. Google ScholarDigital Library
- R. Sabbadin. Possibilistic markov decision process. In ECAI, pages 586--590, 2000.Google Scholar
- W. Stirling, M. Goodrich, and P. Packard. Satisficing equilibria: A non-classical approach to games and decisions. Journal of Autonomous Agents and Multiagent Systems, 5:305--328, 2002. Google ScholarDigital Library
- J. Von-Neumann and O. Morgenstern. The theory of games and economic behavior. In Princeton Univ. Press, 1947.Google Scholar
- K. Wakuta and K. Togawa. Solution procedures for multi-objective markov decison processes. Optimization, 43:29--46, 1998.Google ScholarCross Ref
- D. Wolpert and K. Tumer. Introduction to collective intelligence. Handbook of Agent Technology, AAAI Press/MIT Press, 2000.Google Scholar
- G. Zoltan, Z. Kalmar, and C. Szepesvari. Multi-criteria reinforcement learning. In International Conference on machine Learning (ICML), pages 197--205, 1998. Google ScholarDigital Library
- Towards a formal framework for multi-objective multiagent planning
Recommendations
A robust evolutionary framework for multi-objective optimization
GECCO '08: Proceedings of the 10th annual conference on Genetic and evolutionary computationEvolutionary multi-objective optimization (EMO) methodologies, suggested in the beginning of Nineties, focussed on the task of finding a set of well-converged and well-distributed set of solutions using evolutionary optimization principles. Of the EMO ...
Single-objective and multi-objective formulations of solution selection for hypervolume maximization
GECCO '09: Proceedings of the 11th Annual conference on Genetic and evolutionary computationA new trend in evolutionary multi-objective optimization (EMO) is the handling of a multi-objective problem as an optimization problem of an indicator function. A number of approaches have been proposed under the name of indicator-based evolutionary ...
A Hybrid Framework for Evolutionary Multi-Objective Optimization
Evolutionary multi-objective optimization algorithms are widely used for solving optimization problems with multiple conflicting objectives. However, basic evolutionary multi-objective optimization algorithms have shortcomings, such as slow convergence ...
Comments