Improving Max-Sum through Decimation to Solve Loopy Distributed Constraint Optimization Problems
This work addresses performance issues in solving loopy DCOPs, which is an incremental improvement for distributed constraint optimization.
The authors tackled the problem of poor performance in Max-Sum inference algorithms for loopy distributed constraint optimization problems (DCOP) by proposing DeciMaxSum, a method that incorporates decimation policies. The result showed that some policy combinations outperformed state-of-the-art competitors in empirical evaluations on the Ising model benchmark.
In the context of solving large distributed constraint optimization problems (DCOP), belief-propagation and approximate inference algorithms are candidates of choice. However, in general, when the factor graph is very loopy (i.e. cyclic), these solution methods suffer from bad performance, due to non-convergence and many exchanged messages. As to improve performances of the Max-Sum inference algorithm when solving loopy constraint optimization problems, we propose here to take inspiration from the belief-propagation-guided dec-imation used to solve sparse random graphs (k-satisfiability). We propose the novel DeciMaxSum method, which is parameterized in terms of policies to decide when to trigger decimation, which variables to decimate, and which values to assign to decimated variables. Based on an empirical evaluation on a classical BP benchmark (the Ising model), some of these combinations of policies exhibit better performance than state-of-the-art competitors.