Improving Max-Sum through Decimation to Solve Loopy Distributed Constraint Optimization Problems

Abstract : 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.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01534066
Contributor : Gauthier Picard <>
Submitted on : Wednesday, June 7, 2017 - 10:36:49 AM
Last modification on : Monday, May 6, 2019 - 2:48:44 PM
Long-term archiving on : Friday, September 8, 2017 - 12:23:57 PM

Files

decimation17report.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01534066, version 1
  • ARXIV : 1706.02209

Citation

Jesús Cerquides, Rémi Emonet, Gauthier Picard, Juan A. Rodríguez-Aguilar. Improving Max-Sum through Decimation to Solve Loopy Distributed Constraint Optimization Problems. [Research Report] Ecole des Mines de Saint Etienne (EMSE). 2017. ⟨hal-01534066⟩

Share

Metrics

Record views

753

Files downloads

366