Abstract
A highly influential ingredient of many techniques designed to exploit sparsity in numerical optimization is the so-called chordal extension of a graph representation of the optimization problem. The definitive relation between chordal extension and the performance of the optimization algorithm that uses the extension is not a mathematically understood task. For this reason, we follow the current research trend of looking at Combinatorial Optimization tasks by using a Machine Learning lens, and we devise a framework for learning elimination rules yielding high-quality chordal extensions. As a first building block of the learning framework, we propose an imitation learning scheme that mimics the elimination ordering provided by an expert rule. Results show that our imitation learning approach is effective in learning two classical elimination rules: the minimum degree and minimum fill-in heuristics, using simple Graph Neural Network models with only a handful of parameters. Moreover, the learned policies display remarkable generalization performance, across both graphs of larger size, and graphs from a different distribution.
Similar content being viewed by others
Notes
SuiteSparse matrix collection was formerly known as the University of Florida sparse matrix collection.
References
Agler, J., Helton, W., McCullough, S., Rodman, L.: Positive semidefinite matrices with a given sparsity pattern. Linear Algebra Appl. 107, 101–149 (1988). https://doi.org/10.1016/0024-3795(88)90240-6
Bagnell, J., Chestnutt, J., Bradley, D.M., Ratliff, N.D.: Boosting structured prediction for imitation learning. In: Advances in Neural Information Processing Systems, pp. 1153–1160 (2007)
Bengio, Y., Lodi, A., Prouvost, A.: Machine learning for combinatorial optimization: a methodological tour d’horizon. arXiv preprint arXiv:1811.06128 (2018)
Bergman, D., Cardonha, C.H., Cire, A.A., Raghunathan, A.U.: On the minimum chordal completion polytope. Oper. Res. 67(2), 532–547 (2019). https://doi.org/10.1287/opre.2018.1783
Bixby, R.E.: Solving real-world linear programs: a decade and more of progress. Oper. Res. 50(1), 3–15 (2002). https://doi.org/10.1287/opre.50.1.3.17780
Davis, T.A., Hu, Y.: The university of florida sparse matrix collection. ACM Trans. Math. Softw. 38(1), 1:1–1:25 (2011). https://doi.org/10.1145/2049662.2049663
Duvenaud, D.K., Maclaurin, D., Iparraguirre, J., Bombarell, R., Hirzel, T., Aspuru-Guzik, A., Adams, R.P.: Convolutional networks on graphs for learning molecular fingerprints. In: Advances in Neural Information Processing Systems, pp. 2224–2232 (2015)
Fomin, F.V., Philip, G., Villanger, Y.: Minimum fill-in of sparse graphs: Kernelization and approximation. Algorithmica 71(1), 1–20 (2015). https://doi.org/10.1007/s00453-013-9776-1
Fukuda, M., Kojima, M., Murota, K., Nakata, K.: Exploiting sparsity in semidefinite programming via matrix completion I: general framework. SIAM J. Optim. 11(3), 647–674 (2001)
Garstka, M., Cannon, M., Goulart, P.: A Clique Graph Based Merging Strategy for Decomposable SDPS (2019)
Gasse, M., Chételat, D., Ferroni, N., Charlin, L., Lodi, A.: Exact combinatorial optimization with graph convolutional neural networks. arXiv preprint arXiv:1906.01629 (2019)
George, A.: Nested dissection of a regular finite element mesh. SIAM J. Numer. Anal. 10(2), 345–363 (1973). https://doi.org/10.1137/0710032
George, A., Liu, J.W.: The evolution of the minimum degree ordering algorithm. SIAM Rev. 31(1), 1–19 (1989). https://doi.org/10.1137/1031001
Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp. 249–256 (2010)
Gori, M., Monfardini, G., Scarselli, F.: A new model for learning in graph domains. In: Proceedings of 2005 IEEE International Joint Conference on Neural Networks, 2005., vol. 2, pp. 729–734. IEEE (2005)
Guermouche, A., L’Excellent, J.Y., Utard, G.: Impact of reordering on the memory of a multifrontal solver. Parallel Comput. Parallel Matrix Algorithms Appl. 29(9), 1191–1218 (2003). https://doi.org/10.1016/S0167-8191(03)00099-1
Hamilton, W., Ying, Z., Leskovec, J.: Inductive representation learning on large graphs. In: Advances in Neural Information Processing Systems, pp. 1024–1034 (2017)
Hamilton, W.L., Ying, R., Leskovec, J.: Representation learning on graphs: methods and applications. arXiv preprint arXiv:1709.05584 (2017)
Howard, R.A.: Dynamic programming and Markov processes. (1960)
Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907 (2016)
Kullback, S.: Information theory and statistics. Courier Corporation (1997)
Li, Z., Chen, Q., Koltun, V.: Combinatorial optimization with graph convolutional networks and guided tree search. In: Advances in Neural Information Processing Systems, pp. 539–548 (2018)
Liu, J.W.: The role of elimination trees in sparse factorization. SIAM J. Matrix Anal. Appl. 11(1), 134–172 (1990). https://doi.org/10.1137/0611010
Majumdar, A., Hall, G., Ahmadi, A.A.: A survey of recent scalability improvements for semidefinite programming with applications in machine learning, control, and robotics. arXiv preprint arXiv:1908.05209 (2019)
Markowitz, H.M.: The elimination form of the inverse and its application to linear programming. Manag. Sci. 3(3), 255–269 (1957). https://doi.org/10.1287/mnsc.3.3.255
Nakata, K., Fujisawa, K., Fukuda, M., Kojima, M., Murota, K.: Exploiting sparsity in semidefinite programming via matrix completion II: implementation and numerical results. Math. Progr. 95(2), 303–327 (2003)
Pan, Y., Cheng, C.A., Saigol, K., Lee, K., Yan, X., Theodorou, E., Boots, B.: Agile autonomous driving using end-to-end deep imitation learning. In: Robotics: Science and Systems (2018)
Ratliff, N.D., Silver, D., Bagnell, J.A.: Learning to search: functional gradient techniques for imitation learning. Auton. Robots 27(1), 25–53 (2009)
Rose, D.J.: A graph-theoretic study of the numerical solution of sparse positive definite systems of linear equations. Graph Theory and Computing, pp. 183–217. Elsevier, New York (1972)
Ross, S., Bagnell, D.: Efficient reductions for imitation learning. In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, pp. 661–668 (2010)
Ross, S., Gordon, G., Bagnell, D.: A reduction of imitation learning and structured prediction to no-regret online learning. In: Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, pp. 627–635 (2011)
Rothberg, E., Hendrickson, B.: Sparse matrix ordering methods for interior point linear programming. INFORMS J. Comput. 10(1), 107–113 (1998). https://doi.org/10.1287/ijoc.10.1.107
Schaal, S.: Is imitation learning the route to humanoid robots? Trends Cognit. Sci. 3(6), 233–242 (1999)
Silver, D., Bagnell, J., Stentz, A.: High performance outdoor navigation from overhead data using imitation learning. Robotics: Science and Systems IV, Zurich, Switzerland (2008)
Sliwak, J., Anjos, M., Létocart, L., Maeght, J., Traversi, E.: Improving clique decompositions of semidefinite relaxations for optimal power flow problems. EasyChair Preprint no. 2546 (EasyChair, 2020)
Vandenberghe, L., Andersen, M.S., et al.: Chordal graphs and semidefinite optimization. Found. Trends® Optim. 1(4), 241–433 (2015)
Vanderbei, R.J.: LOQO: an interior point code for quadratic programming. Optim. Methods Softw. 11(1–4), 451–484 (1999). https://doi.org/10.1080/10556789908805759
Wright, S.: Primal-dual interior-point methods. Soc. Ind. Appl. Math. (1997). https://doi.org/10.1137/1.9781611971453
Yannakakis, M.: Computing the minimum fill-in is NP-complete. SIAM J. Algebraic Discrete Methods 2(1), 77–79 (1981). https://doi.org/10.1137/0602010
Ying, Z., Bourgeois, D., You, J., Zitnik, M., Leskovec, J.: Gnnexplainer: Generating explanations for graph neural networks. In: H. Wallach, H. Larochelle, A. Beygelzimer, F. d’Alché Buc, E. Fox, R. Garnett (eds.) Advances in Neural Information Processing Systems 32, pp. 9244–9255. Curran Associates, Inc. (2019)
Zheng, Y., Fantuzzi, G.: Sum-of-squares chordal decomposition of polynomial matrix inequalities. arXiv:2007.11410 [cs, eess, math] (2020)
Zheng, Y., Fantuzzi, G., Papachristodoulou, A., Goulart, P., Wynn, A.: Chordal decomposition in operator-splitting methods for sparse semidefinite programs. Math. Progr. (2019). https://doi.org/10.1007/s10107-019-01366-3
Acknowledgements
We are indebted to three anonymous referees for their careful reading and constructive suggestions that helped us improving the quality and readability of the paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Liu, D., Lodi, A. & Tanneau, M. Learning chordal extensions. J Glob Optim 81, 3–22 (2021). https://doi.org/10.1007/s10898-020-00973-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10898-020-00973-1