PKOPT: Faster k-Optimal Solution for DCOP by Improving Group Selection Strategy

Abstract

A significant body of work in multiagent systems over more than two decades has focused on multi-agent coordination. Many challenges in multi-agent coordination can be modeled as Distributed Constraint Optimizations (DCOPs). Many complete and incomplete algorithms have been introduced for DCOPs but complete algorithms are often impractical for large-scale and dynamic environments which lead to study incomplete algorithms. Some incomplete algorithms produce k-optimal solutions; a k-optimal solution is the one that cannot be improved by any deviation by k or fewer agents. In this paper we focus on the only k-optimal algorithm which works for arbitrary k, entitled as KOPT. In both complete and incomplete algorithms, computational complexity is the major concern. Different approaches are introduced to solve this problem and improve existing algorithms. The main contribution of this paper is to decrease computational complexity of KOPT algorithm by introducing a new method for selecting leaders which should assign new values to a group of agents. This new approach is called Partial KOPT (PKOPT). PKOPT is an effective method to reduce computational load and power consumption in implementation. This paper under various assumptions presents an analysis of sequential and stochastic PKOPT algorithms.