Distributed Bayesian: A Continuous Distributed Constraint Optimization Problem Solver

In this paper, the novel Distributed Bayesian (D-Bay) algorithm is presented for solving multi-agent problems within Continuous Constraint Optimization Problem (C-DCOP) framework. This framework extends classical DCOP towards utility functions with continuous domains. D-Bay solves a C-DCOP by utilizing optimization adaptive sampling of variables. We theoretically show that converges to global optimum Lipschitz functions. The performance evaluated empirically based on sample efficiency. proposed compared state-of-the-art and solvers. generates better solutions while requiring fewer samples.