Abstract
Max-sum is a version of Belief Propagation, used for solving DCOPs. On tree-structured problems, Max-sum converges to the optimal solution in linear time. When the constraint graph representing the problem includes multiple cycles, Max-sum might not converge and explore low quality solutions. Damping is a method that increases the chances that Max-sum will converge. Damped Max-sum (DMS) was recently found to produce high quality solutions for DCOP when combined with an anytime framework.
We propose a novel method for adjusting the level of asymmetry in the factor graph, in order to achieve a balance between exploitation and exploration, when using Max-sum for solving DCOPs. By converting a standard factor graph to an equivalent split constraint factor graph (SCFG), in which each function-node is split to two function-nodes, we can control the level of asymmetry for each constraint. Our empirical results demonstrate that by applying DMS to SCFGs with a minor level of asymmetry we can find high quality solutions in a small number of iterations, even without using an anytime framework. As part of our investigation of this success, we prove that for a factor-graph with a single constraint, if this constraint is split symmetrically, Max-sum applied to the resulting cycle is guaranteed to converge to the optimal solution and demonstrate that for an asymmetric split, convergence is not guaranteed.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
A similar factor graph was used in [31] for representing asymmetric DCOPs.
- 2.
The inference algorithm for minimization problems is actually Min-sum. However, we will continue to refer to it as Max-sum since this name is widely accepted.
- 3.
We say that a variable is involved in a constraint if it is one of the variables the constraint refers to.
- 4.
For an example of the need to break ties in the factor-graph see [30].
- 5.
- 6.
- 7.
Recall that we assumed in Sect. 3.1 that there are no ties, so such a cost is unique.
- 8.
Without loss of generality, we assume that k is odd. If it was even, then the assumption was that the cost is \(\frac{k}{2}c\).
- 9.
further insights on the relation between the success of our empirical results and the properties presented in this section are detailed in Sect. 6.2.
- 10.
This range was selected so that the numbers do not become too small and due to precision, generate distorted SCFGs. Obviously, if the input costs are between 0 and 100, adding 100 to each cost can be the first step of the splitting method.
- 11.
For lack of space we do not present convergence graphs for the other problems. As expected the meeting scheduling convergence results were similar to graph coloring while the results for the other problem types were similar to the convergence results of the sparse uniform random problems.
References
Chli, M., Winsper, M.: Using the max-sum algorithm for supply chain emergence in dynamic multiunit environments. IEEE Trans. Syst. Man Cybern. 45(3), 422–435 (2015)
Cohen, L., Zivan, R.: Max-sum revisited: the real power of damping. In: Sukthankar, G., Rodriguez-Aguilar, J.A. (eds.) AAMAS 2017. LNCS (LNAI), vol. 10643, pp. 111–124. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-71679-4_8
Farinelli, A., Rogers, A., Petcu, A., Jennings, N.: Decentralised coordination of low-power embedded devices using the max-sum algorithm. In: Proceedings of the 7th International Joint Conference on Autonomous Agents and Multiagent Systems, vol. 2, pp. 639–646. International Foundation for Autonomous Agents and Multiagent Systems (2008)
Farinelli, A., Rogers, A., Petcu, A., Jennings, N.R.: Decentralized coordination of low-power embedded devices using the max-sum algorithm. In: AAMAS, pp. 639–646 (2008)
Farinelli, A., Rogers, A., Jennings, N.R.: Agent-based decentralised coordination for sensor networks using the max-sum algorithm. Auton. Agents Multi-Agent Syst. 28(3), 337–380 (2014)
Gershman, A., Meisels, A., Zivan, R.: Asynchronous forward bounding. J. Artif. Intell. Res. 34, 25–46 (2009)
Kschischang, F.R., Frey, B.J., Loeliger, H.A.: Factor graphs and the sum-product algorithm. IEEE Trans. Inf. Theory 47(2), 181–208 (2001)
Lazic, N., Frey, B., Aarabi, P.: Solving the uncapacitated facility location problem using message passing algorithms. In: International Conference on Artificial Intelligence and Statistics, pp. 429–436 (2010)
Macarthur, K.S., Stranders, R., Ramchurn, S.D., Jennings, N.R.: A distributed anytime algorithm for dynamic task allocation in multi-agent systems. In: AAAI (2011)
Modi, P.J., Shen, W., Tambe, M., Yokoo, M.: Adopt: asynchronous distributed constraints optimizationwith quality guarantees. Artif. Intell. 161(1–2), 149–180 (2005)
Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, San Francisco (1988)
Petcu, A., Faltings, B.: A scalable method for multiagent constraint optimization. In: IJCAI, pp. 266–271 (2005)
Petcu, A., Faltings, B.: Approximations in distributed optimization. In: van Beek, P. (ed.) CP 2005. LNCS, vol. 3709, pp. 802–806. Springer, Heidelberg (2005). https://doi.org/10.1007/11564751_68
Pretti, M.: A message-passing algorithm with damping. J. Stat. Mech. Theory Exp. 11, P11008 (2005)
Pujol-Gonzalez, M., Cerquides, J., Meseguer, P., RodrÃguez-Aguilar, J.A., Tambe, M.: Engineering the decentralized coordination of UAVs with limited communication range. In: Bielza, C., et al. (eds.) CAEPIA 2013. LNCS (LNAI), vol. 8109, pp. 199–208. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-40643-0_21
Ramchurn, S.D., Farinelli, A., Macarthur, K.S., Jennings, N.R.: Decentralized coordination in robocup rescue. Comput. J. 53(9), 1447–1461 (2010)
Rogers, A., Farinelli, A., Stranders, R., Jennings, N.R.: Bounded approximate decentralized coordination via the max-sum algorithm. Artif. Intell. 175(2), 730–759 (2011)
Ruozzi, N., Tatikonda, S.: Message-passing algorithms: reparameterizations and splittings. IEEE Trans. Inf. Theory 59(9), 5860–5881 (2013)
Som, P., Chockalingam, A.: Damped belief propagation based near-optimal equalization of severely delay-spread UWB MIMO-ISI channels. In: 2010 IEEE International Conference on Communications (ICC), pp. 1–5. IEEE (2010)
Stranders, R., Farinelli, A., Rogers, A., Jennings, N.R.: Decentralised coordination of mobile sensors using the max-sum algorithm. In: IJCAI 2009, Proceedings of the 21st International Joint Conference on Artificial Intelligence, Pasadena, California, USA, 11–17 July 2009, pp. 299–304 (2009)
Tarlow, D., Givoni, I., Zemel, R., Frey, B.: Graph cuts is a max-product algorithm. In: Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence (2011)
Tarlow, D., Givoni, I., Zemel, R.: Hop-map: efficient message passing with high order potentials. In: AISTATS, vol. 9, pp. 812–819 (2010)
Teacy, W.T.L., Farinelli, A., Grabham, N.J., Padhy, P., Rogers, A., Jennings, N.R.: Max-sum decentralized coordination for sensor systems. In: AAMAS, pp. 1697–1698 (2008)
Weiss, Y.: Correctness of local probability propagation in graphical models with loops. Neural Comput. 12(1), 1–41 (2000)
Yanover, C., Meltzer, T., Weiss, Y.: Linear programming relaxations and belief propagation - an empirical study. J. Mach. Learn. Res. 7, 1887–1907 (2006)
Yedidsion, H., Zivan, R., Farinelli, A.: Explorative max-sum for teams of mobile sensing agents. In: International Conference on Autonomous Agents and Multi-Agent Systems, AAMAS 2014, Paris, France, 5–9 May 2014, pp. 549–556 (2014)
Yeoh, W., Felner, A., Koenig, S.: Bnb-ADOPT: an asynchronous branch-and-bound DCOP algorithm. Artif. Intell. Res. (JAIR) 38, 85–133 (2010)
Zhang, W., Xing, Z., Wang, G., Wittenburg, L.: Distributed stochastic search and distributed breakout: properties, comparison and applications to constraints optimization problems in sensor networks. Artif. Intell. 161(1–2), 55–88 (2005)
Zivan, R., Okamoto, S., Peled, H.: Explorative anytime local search for distributed constraint optimization. Artif. Intell. 211, 1–21 (2014)
Zivan, R., Parash, T., Cohen, L., Peled, H., Okamoto, S.: Balancing exploration and exploitation in incomplete min/max-sum inference for distributed constraint optimization. Auton. Agents Multi-Agent Syst. 31(5), 1165–1207 (2017)
Zivan, R., Parash, T., Naveh, Y.: Applying max-sum to asymmetric distributed constraint optimization. In: Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI 2015, Buenos Aires, Argentina, 25–31 July 2015, pp. 432–439 (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Cohen, L., Zivan, R. (2018). Balancing Asymmetry in Max-sum Using Split Constraint Factor Graphs. In: Hooker, J. (eds) Principles and Practice of Constraint Programming. CP 2018. Lecture Notes in Computer Science(), vol 11008. Springer, Cham. https://doi.org/10.1007/978-3-319-98334-9_43
Download citation
DOI: https://doi.org/10.1007/978-3-319-98334-9_43
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-98333-2
Online ISBN: 978-3-319-98334-9
eBook Packages: Computer ScienceComputer Science (R0)