Abstract
Which is the minimum number of variables that need branching for a given MIP instance? Can this information be effective in producing compact branching trees, hence improving the performance of a state-of-the-art solver? In this paper we present a restart exact MIP solution scheme where a set covering model is used to find a small set of variables (a “backdoor”, in the terminology of [8]) to be used as first-choice variables for branching. In a preliminary “sampling” phase, our method quickly collects a number of relevant low-cost fractional solutions that qualify as obstacles for LP bound improvement. Then a set covering model is solved to detect a small subset of variables (the backdoor) that “cover the fractionality” of the collected fractional solutions. These backdoor variables are put in a priority branching list, and a black-box MIP solver is eventually run—in its default mode—by taking this list into account, thus avoiding any other interference with its highly-optimized internal mechanisms. Computational results on a large set of instances from MIPLIB 2010 are presented, showing that some speedup can be achieved even with respect to a state-of-the-art solver such as IBM ILOG Cplex 12.2.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Achterberg, T.: Constraint Integer Programming. PhD thesis, Technische Universität Berlin; Fakultät II - Mathematik und Naturwissenschaften. Institut für Mathematik (2007)
Balas, E., Ceria, S., Cornuéjols, G., Natraj, N.: Gomory cuts revisited. Operations Research Letters 19, 1–9 (1996)
Bénichou, M., Gauthier, J.M., Girodet, P., Hentges, G., Ribière, G., Vincent, O.: Experiments in mixed integer linear programming. Mathematical Programming 1, 76–94 (1971)
Chvátal, V.: Resolution search. DAMATH: Discrete Applied Mathematics and Combinatorial Operations Research and Computer Science 73 (1997)
Dilkina, B., Gomes, C.P., Malitsky, Y., Sabharwal, A., Sellmann, M.: Backdoors to Combinatorial Optimization: Feasibility and Optimality. In: van Hoeve, W.-J., Hooker, J.N. (eds.) CPAIOR 2009. LNCS, vol. 5547, pp. 56–70. Springer, Heidelberg (2009)
Karzan, F.K., Nemhauser, G.L., Savelsbergh, M.W.P.: Information-based branching schemes for binary linear mixed integer problems. Mathematical Programming Computation 1, 249–293 (2009)
MIPLIB 2010. Preliminary version, http://miplib.zip.de/
Williams, R., Gomes, C.P., Selman, B.: Backdoors to typical case complexity. In: Gottlob, G., Walsh, T. (eds.) IJCAI 2003: Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, pp. 1173–1178. Morgan Kaufmann, San Francisco (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Fischetti, M., Monaci, M. (2011). Backdoor Branching. In: Günlük, O., Woeginger, G.J. (eds) Integer Programming and Combinatoral Optimization. IPCO 2011. Lecture Notes in Computer Science, vol 6655. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-20807-2_15
Download citation
DOI: https://doi.org/10.1007/978-3-642-20807-2_15
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-20806-5
Online ISBN: 978-3-642-20807-2
eBook Packages: Computer ScienceComputer Science (R0)