Abstract
Algorithm selection refers to the task of automatically selecting the most suitable algorithm for solving an instance of a computational problem from a set of candidate algorithms. Here, suitability is typically measured in terms of the algorithms’ runtimes. To allow the selection of algorithms on new problem instances, machine learning models are trained on previously observed performance data and then used to predict the algorithms’ performances. Due to the computational effort, the execution of such algorithms is often prematurely terminated, which leads to right-censored observations representing a lower bound on the actual runtime. While simply neglecting these censored samples leads to overly optimistic models, imputing them with precise though hypothetical values, such as the commonly used penalized average runtime, is a rather arbitrary and biased approach. In this paper, we propose a simple regression method based on so-called superset learning, in which right-censored runtime data are explicitly incorporated in terms of interval-valued observations, offering an intuitive and efficient approach to handling censored data. Benchmarking on publicly available algorithm performance data, we demonstrate that it outperforms the aforementioned naïve ways of dealing with censored samples and is competitive to established methods for censored regression in the field of algorithm selection.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Here, we make the assumptions that \(L(y,y)=0\) for all \(y \in \mathcal {Y}\), \(L(y, \cdot )\) is monotone decreasing on \((-\infty , y)\) and monotone increasing on \((y, \infty )\), which hold for all reasonable loss functions.
- 2.
- 3.
Note that for the TSP-LION2015 scenario, only 9 out of 10 folds were evaluated using the Schmee&Hahn imputation due to technical issues.
References
Amadini, R., Gabbrielli, M., Mauro, J.: SUNNY: a lazy portfolio approach for constraint solving. Theor. Pract. Log. Prog. 14(4–5), 509–524 (2014)
Bischl, B., et al.: ASlib: a benchmark library for algorithm selection. Artif. Intell. 237, 41–58 (2016)
Cabannes, V., Rudi, A., Bach, F.: Structured prediction with partial labelling through the infimum loss. In: Proceedings of ICML, International Conference on Machine Learning (2020)
Eggensperger, K., Lindauer, M., Hoos, H.H., Hutter, F., Leyton-Brown, K.: Efficient benchmarking of algorithm configurators via model-based surrogates. Mach. Learn. 107(1), 15–41 (2017). https://doi.org/10.1007/s10994-017-5683-z
Gagliolo, M., Legrand, C.: Algorithm survival analysis. In: Bartz-Beielstein, T., Chiarandini, M., Paquete, L., Preuss, M. (eds.) Experimental Methods for the Analysis of Optimization Algorithms, pp. 161–184. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-02538-9_7
Gagliolo, M., Schmidhuber, J.: Leaning dynamic algorithm portfolios. Ann. Math. Artif. Intell. 47, 295–328 (2006)
Gomes, C.P., Selman, B., Crato, N.: Heavy-tailed distributions in combinatorial search. In: Smolka, G. (ed.) CP 1997. LNCS, vol. 1330, pp. 121–135. Springer, Heidelberg (1997). https://doi.org/10.1007/BFb0017434
Hanselle, J., Tornede, A., Wever, M., Hüllermeier, E.: Hybrid ranking and regression for algorithm selection. In: Schmid, U., Klügl, F., Wolter, D. (eds.) KI 2020. LNCS (LNAI), vol. 12325, pp. 59–72. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58285-2_5
Hüllermeier, E., Cheng, W.: Superset learning based on generalized loss minimization. In: Appice, A., Rodrigues, P.P., Santos Costa, V., Gama, J., Jorge, A., Soares, C. (eds.) ECML PKDD 2015. LNCS (LNAI), vol. 9285, pp. 260–275. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-23525-7_16
Hutter, F., Hoos, H.H., Leyton-Brown, K.: Bayesian optimization with censored response data. In: NIPS Workshop on Bayesian Optimization, Sequential Experimental Design, and Bandits (2011)
Hüllermeier, E.: Learning from imprecise and fuzzy observations: data disambiguation through generalized loss minimization. Int. J. Approx. Reason. 55(7), 1519–1534 (2014). Special issue: Harnessing the information contained in low-quality data sources
Kadioglu, S., Malitsky, Y., Sellmann, M., Tierney, K.: ISAC - instance-specific algorithm configuration. In: ECAI (2010)
Kerschke, P., Hoos, H.H., Neumann, F., Trautmann, H.: Automated algorithm selection: survey and perspectives. Evol. Comput. 27(1), 3–45 (2019)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, 7–9 May 2015. Conference Track Proceedings (2015)
Kleinbaum, D.G., Klein, M.: Survival Analysis. Survival Analysis, vol. 3. Springer, New York (2012). https://doi.org/10.1007/978-1-4419-6646-9
Kotthoff, L.: Hybrid regression-classification models for algorithm selection. In: ECAI, pp. 480–485 (2012)
Lindauer, M., Bergdoll, R.-D., Hutter, F.: An empirical study of per-instance algorithm scheduling. In: Festa, P., Sellmann, M., Vanschoren, J. (eds.) LION 2016. LNCS, vol. 10079, pp. 253–259. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-50349-3_20
Lobjois, L., Lemaître, M., et al.: Branch and bound algorithm selection by performance prediction. In: AAAI/IAAI, pp. 353–358 (1998)
Pihera, J., Musliu, N.: Application of machine learning to algorithm selection for TSP. In: 26th IEEE International Conference on Tools with Artificial Intelligence, ICTAI 2014, Limassol, Cyprus, 10–12 November 2014, pp. 47–54. IEEE Computer Society (2014)
Rice, J.R.: The algorithm selection problem. Adv. Comput. 15, 65–118 (1976)
Schmee, J., Hahn, G.J.: A simple method for regression analysis with censored data. Technometrics 21(4), 417–432 (1979)
Tornede, A., Wever, M., Hüllermeier, E.: Extreme algorithm selection with dyadic feature representation. In: Appice, A., Tsoumakas, G., Manolopoulos, Y., Matwin, S. (eds.) DS 2020. LNCS (LNAI), vol. 12323, pp. 309–324. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-61527-7_21
Tornede, A., Wever, M., Hüllermeier, E.: Towards meta-algorithm selection. In: Workshop on Meta-Learning (MetaLearn 2020) @ NeurIPS 2020 (2020)
Tornede, A., Wever, M., Werner, S., Mohr, F., Hüllermeier, E.: Run2survive: a decision-theoretic approach to algorithm selection based on survival analysis. In: Asian Conference on Machine Learning, pp. 737–752. PMLR (2020)
Xu, L., Hutter, F., Hoos, H., Leyton-Brown, K.: Hydra-mip: automated algorithm configuration and selection for mixed integer programming. In: RCRA Workshop @ IJCAI (2011)
Xu, L., Hutter, F., Hoos, H.H., Leyton-Brown, K.: The design and analysis of an algorithm portfolio for SAT. In: Bessière, C. (ed.) CP 2007. LNCS, vol. 4741, pp. 712–727. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-74970-7_50
Acknowledgments
This work was supported by the German Federal Ministry of Economic Affairs and Energy (BMWi) within the “Innovationswettbewerb Künstliche Intelligenz” and the German Research Foundation (DFG) within the Collaborative Research Center “On-The-Fly Computing” (SFB 901/3 project no. 160364472). The authors also gratefully acknowledge support of this project through computing time provided by the Paderborn Center for Parallel Computing (PC\(^2\)).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Hanselle, J., Tornede, A., Wever, M., Hüllermeier, E. (2021). Algorithm Selection as Superset Learning: Constructing Algorithm Selectors from Imprecise Performance Data. In: Karlapalem, K., et al. Advances in Knowledge Discovery and Data Mining. PAKDD 2021. Lecture Notes in Computer Science(), vol 12712. Springer, Cham. https://doi.org/10.1007/978-3-030-75762-5_13
Download citation
DOI: https://doi.org/10.1007/978-3-030-75762-5_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-75761-8
Online ISBN: 978-3-030-75762-5
eBook Packages: Computer ScienceComputer Science (R0)