Advertisement

Deep Learning as a Competitive Feature-Free Approach for Automated Algorithm Selection on the Traveling Salesperson Problem

  • Moritz SeilerEmail author
  • Janina Pohl
  • Jakob Bossek
  • Pascal Kerschke
  • Heike Trautmann
Conference paper
  • 329 Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12269)

Abstract

In this work we focus on the well-known Euclidean Traveling Salesperson Problem (TSP) and two highly competitive inexact heuristic TSP solvers, EAX and LKH, in the context of per-instance algorithm selection (AS). We evolve instances with \(1\,000\) nodes where the solvers show strongly different performance profiles. These instances serve as a basis for an exploratory study on the identification of well-discriminating problem characteristics (features). Our results in a nutshell: we show that even though (1) promising features exist, (2) these are in line with previous results from the literature, and (3) models trained with these features are more accurate than models adopting sophisticated feature selection methods, the advantage is not close to the virtual best solver in terms of penalized average runtime and so is the performance gain over the single best solver. However, we show that a feature-free deep neural network based approach solely based on visual representation of the instances already matches classical AS model results and thus shows huge potential for future studies.

Keywords

Automated algorithm selection Traveling Salesperson Problem Feature-based approaches Deep learning 

Notes

Acknowledgments

The authors acknowledge support by the European Research Center for Information Systems (ERCIS) .

References

  1. 1.
    Alissa, M., Sim, K., Hart, E.: Algorithm selection using deep learning without feature extraction. In: Proceedings of the Genetic and Evolutionary Computation Conference GECCO 2019, pp. 198–206. Association for Computing Machinery, New York (2019).  https://doi.org/10.1145/3321707.3321845
  2. 2.
    Bischl, B., et al.: ASlib: a benchmark library for algorithm selection. Artif. Intell. 237, 41–58 (2016).  https://doi.org/10.1016/j.artint.2016.04.003MathSciNetCrossRefzbMATHGoogle Scholar
  3. 3.
    Bischl, B., et al.: mlr: machine learning in R. J. Mach. Learn. Res. (JMLR) 17(170), 1–5 (2016). http://jmlr.org/papers/v17/15-066.htmlMathSciNetzbMATHGoogle Scholar
  4. 4.
    Bischl, B., Mersmann, O., Trautmann, H., Preuss, M.: Algorithm selection based on exploratory landscape analysis and cost-sensitive learning. In: Proceedings of the 14th Annual Conference on Genetic and Evolutionary Computation (GECCO), pp. 313–320. ACM, July 2012.  https://doi.org/10.1145/2330163.2330209. http://dl.acm.org/citation.cfm?doid=2330163.2330209
  5. 5.
    Bossek, J.: Salesperson: computation of instance features and R interface to the state-of-the-art exact and inexact solvers for the traveling salesperson problem (2017). https://github.com/jakobbossek/salesperson. R package version 1.0.0
  6. 6.
    Bossek, J., Kerschke, P., Neumann, A., Wagner, M., Neumann, F., Trautmann, H.: Evolving diverse TSP instances by means of novel and creative mutation operators. In: Friedrich, T., Doerr, C., Arnold, D. (eds.) Proceedings of the 15th ACM/SIGEVO Workshop on Foundations of Genetic Algorithms (FOGA XV), pp. 58–71. ACM, Potsdam (2019)Google Scholar
  7. 7.
    Bossek, J., Trautmann, H.: Evolving instances for maximizing performance differences of state-of-the-art inexact TSP solvers. In: Festa, P., Sellmann, M., Vanschoren, J. (eds.) LION 2016. LNCS, vol. 10079, pp. 48–59. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-50349-3_4CrossRefGoogle Scholar
  8. 8.
    Bossek, J., Trautmann, H.: Understanding characteristics of evolved instances for state-of-the-art inexact TSP solvers with maximum performance difference. In: Adorni, G., Cagnoni, S., Gori, M., Maratea, M. (eds.) AI*IA 2016. LNCS (LNAI), vol. 10037, pp. 3–12. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-49130-1_1CrossRefGoogle Scholar
  9. 9.
    Chen, T., et al.: XGBoost: extreme gradient boosting (2019). https://CRAN.R-project.org/package=xgboost. R package version 0.90.0.2
  10. 10.
    Glorot, X., Bordes, A., Bengio, Y.: Deep sparse rectifier neural networks. In: Gordon, G.J., Dunson, D.B., Dudík, M. (eds.) Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, AISTATS 2011, Fort Lauderdale, USA, 11–13 April 2011. JMLR Proceedings, vol. 15, pp. 315–323. JMLR.org (2011). http://proceedings.mlr.press/v15/glorot11a/glorot11a.pdf
  11. 11.
    Guyon, I., Elisseeff, A.: An introduction to feature extraction. In: Guyon, I., Nikravesh, M., Gunn, S., Zadeh, L.A. (eds.) Feature Extraction. STUDFUZZ, vol. 207, pp. 1–25. Springer, Heidelberg (2006).  https://doi.org/10.1007/978-3-540-35488-8_1CrossRefGoogle Scholar
  12. 12.
    Härdle, W.K., Simar, L.: Applied Multivariate Statistical Analysis, 4th edn. Springer, Heidelberg (2015).  https://doi.org/10.1007/978-3-662-45171-7CrossRefzbMATHGoogle Scholar
  13. 13.
    Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer, Heidelberg (2009). http://www.springer.com/de/book/9780387848570CrossRefGoogle Scholar
  14. 14.
    Helsgaun, K.: An effective implementation of the lin-kernighan traveling salesman heuristic. Eur. J. Oper. Res. 126(1), 106–130 (2000)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Hutter, F., Xu, L., Hoos, H.H., Leyton-Brown, K.: Algorithm runtime prediction: methods & evaluation. Artif. Intell. 206, 79–111 (2014).  https://doi.org/10.1016/j.artint.2013.10.003MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Hutter, F., Xu, L., Hoos, H.H., Leyton-Brown, K.: Algorithm runtime prediction: methods & evaluation. Artif. Intell. J. (AIJ) 206, 79–111 (2014). http://www.sciencedirect.com/science/article/pii/S0004370213001082MathSciNetCrossRefGoogle Scholar
  17. 17.
    Karatzoglou, A., Smola, A., Hornik, K., Zeileis, A.: kernlab - An S4 package for kernel methods in R. J. Stat. Softw. (JSS) 11(9), 1–20 (2004). http://www.jstatsoft.org/v11/i09/Google Scholar
  18. 18.
    Kerschke, P., Bossek, J., Trautmann, H.: Parameterization of state-of-the-art performance indicators: a robustness study based on inexact TSP solvers. In: Proceedings of the 20th Genetic and Evolutionary Computation Conference (GECCO) Companion, pp. 1737–1744. ACM, Kyoto (2018).  https://doi.org/10.1145/3205651.3208233. http://doi.acm.org/10.1145/3205651.3208233
  19. 19.
    Kerschke, P., Hoos, H.H., Neumann, F., Trautmann, H.: Automated algorithm selection: survey and perspectives. Evol. Comput. (ECJ) 27(1), 3–45 (2019)CrossRefGoogle Scholar
  20. 20.
    Kerschke, P., Kotthoff, L., Bossek, J., Hoos, H.H., Trautmann, H.: Leveraging TSP solver complementarity through machine learning. Evol. Comput. (ECJ) 26(4), 597–620 (2018)CrossRefGoogle Scholar
  21. 21.
    Kerschke, P., Trautmann, H.: Automated algorithm selection on continuous black-box problems by combining exploratory landscape analysis and machine learning. Evol. Comput. 27(1), 99–127 (2019).  https://doi.org/10.1162/evco_a_00236. pMID: 30365386CrossRefGoogle Scholar
  22. 22.
    Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Bengio, Y., LeCun, Y. (eds.) 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, 7–9 May 2015, Conference Track Proceedings (2015). http://arxiv.org/abs/1412.6980
  23. 23.
    Kohavi, R., John, G.H., et al.: Wrappers for feature subset selection. Artif. Intell. 97(1–2), 273–324 (1997)CrossRefGoogle Scholar
  24. 24.
    Kotthoff, L.: Algorithm selection for combinatorial search problems: a survey. AI Mag. 35(3), 48–60 (2014).  https://doi.org/10.1609/aimag.v35i3.2460. https://aaai.org/ojs/index.php/aimagazine/article/view/2460CrossRefGoogle Scholar
  25. 25.
    Kotthoff, L., Kerschke, P., Hoos, H., Trautmann, H.: Improving the state of the art in inexact TSP solving using per-instance algorithm selection. In: Dhaenens, C., Jourdan, L., Marmion, M.-E. (eds.) LION 2015. LNCS, vol. 8994, pp. 202–217. Springer, Cham (2015).  https://doi.org/10.1007/978-3-319-19084-6_18CrossRefGoogle Scholar
  26. 26.
    LeCun, Y., Bengio, Y., et al.: Convolutional networks for images, speech, and time series. In: The Handbook of Brain Theory and Neural Networks, vol. 3361, no. 10, p. 1995 (1995)Google Scholar
  27. 27.
    Liaw, A., Wiener, M.: Classification and regression by randomForest. R News 2(3), 18–22 (2002). https://cran.r-project.org/doc/Rnews/Rnews2002-3.pdfGoogle Scholar
  28. 28.
    Lin, M., Chen, Q., Yan, S.: Network in network. In: Bengio, Y., LeCun, Y. (eds.) 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, 14–16 April 2014, Conference Track Proceedings (2014). http://arxiv.org/abs/1312.4400
  29. 29.
    Lindauer, T.M., Hoos, H.H., Hutter, F., Schaub, T.: AutoFolio: an automatically configured algorithm selector (extended abstract). In: Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), pp. 5025–5029, August 2017.  https://doi.org/10.24963/ijcai.2017/715. https://www.ijcai.org/proceedings/2017/715
  30. 30.
    Malitsky, Y., Sabharwal, A., Samulowitz, H., Sellmann, M.: Algorithm portfolios based on cost-sensitive hierarchical clustering. In: Rossi, F. (ed.) Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence (IJCAI), vol. 13, pp. 608–614. Association for the Advancement of Artificial Intelligence (AAAI), August 2013. https://www.aaai.org/ocs/index.php/IJCAI/IJCAI13/paper/view/6946
  31. 31.
    Mann, H.B., Whitney, D.R.: On a test of whether one of two random variables is stochastically larger than the other. Ann. Math. Stat. 18(1), 50–60 (1947).  https://doi.org/10.1214/aoms/1177730491MathSciNetCrossRefzbMATHGoogle Scholar
  32. 32.
    Mannor, S., Peleg, D., Rubinstein, R.Y.: The cross entropy method for classification. In: Raedt, L.D., Wrobel, S. (eds.) Machine Learning, Proceedings of the Twenty-Second International Conference (ICML 2005), Bonn, Germany, 7–11 August 2005. ACM International Conference Proceeding Series, vol. 119, pp. 561–568. ACM (2005).  https://doi.org/10.1145/1102351.1102422
  33. 33.
    Mersmann, O., Bischl, B., Bossek, J., Trautmann, H., Wagner, M., Neumann, F.: Local search and the traveling salesman problem: a feature-based characterization of problem hardness. In: Hamadi, Y., Schoenauer, M. (eds.) LION 2012. LNCS, pp. 115–129. Springer, Heidelberg (2012).  https://doi.org/10.1007/978-3-642-34413-8_9CrossRefGoogle Scholar
  34. 34.
    Mersmann, O., Bischl, B., Trautmann, H., Wagner, M., Bossek, J., Neumann, F.: A novel feature-based approach to characterize algorithm performance for the traveling salesperson problem. Ann. Math. Artif. Intell. 69(2), 151–182 (2013).  https://doi.org/10.1007/s10472-013-9341-2. https://link.springer.com/article/10.1007/s10472-013-9341-2MathSciNetCrossRefzbMATHGoogle Scholar
  35. 35.
    Nagata, Y., Kobayashi, S.: A powerful genetic algorithm using edge assembly crossover for the traveling salesman problem. INFORMS J. Comput. 25(2), 346–363 (2013)MathSciNetCrossRefGoogle Scholar
  36. 36.
    Peng, H., Long, F., Ding, C.: Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans. Pattern Anal. Mach. Intell. (TPAMI) 27(8), 1226–1238 (2005).  https://doi.org/10.1109/TPAMI.2005.159. https://ieeexplore.ieee.org/abstract/document/1453511CrossRefGoogle Scholar
  37. 37.
    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)Google Scholar
  38. 38.
    Reinelt, G.: TSPLIB-a traveling salesman problem library. ORSA J. Comput. 3(4), 376–384 (1991)CrossRefGoogle Scholar
  39. 39.
    Rice, J.R.: The algorithm selection problem. Adv. Comput. 15, 65–118 (1976). http://www.sciencedirect.com/science/article/pii/S0065245808605203CrossRefGoogle Scholar
  40. 40.
    Rizzini, M., Fawcett, C., Vallati, M., Gerevini, A.E., Hoos, H.H.: Static and dynamic portfolio methods for optimal planning: an empirical analysis. Int. J. Artif. Intell. Tools 26(01), 1–27 (2017).  https://doi.org/10.1142/S0218213017600065. https://www.worldscientific.com/doi/abs/10.1142/S0218213017600065CrossRefGoogle Scholar
  41. 41.
    Robbins, H., Monro, S.: A stochastic approximation method. Ann. Math. Stat. 22, 400–407 (1951)MathSciNetCrossRefGoogle Scholar
  42. 42.
    Ross, P., Schulenburg, S., Marín-Bläzquez, J.G., Hart, E.: Hyper-heuristics: learning to combine simple heuristics in bin-packing problems. In: Proceedings of the 4th Annual Conference on Genetic and Evolutionary Computation GECCO 2002, pp. 942–948. Morgan Kaufmann Publishers Inc., San Francisco (2002)Google Scholar
  43. 43.
    Sim, K., Hart, E., Paechter, B.: A hyper-heuristic classifier for one dimensional bin packing problems: improving classification accuracy by attribute evolution. In: Coello, C.A.C., Cutello, V., Deb, K., Forrest, S., Nicosia, G., Pavone, M. (eds.) PPSN 2012. LNCS, vol. 7492, pp. 348–357. Springer, Heidelberg (2012).  https://doi.org/10.1007/978-3-642-32964-7_35CrossRefGoogle Scholar
  44. 44.
    Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014)MathSciNetzbMATHGoogle Scholar
  45. 45.
    Therneau, T., Atkinson, B.: rpart: recursive partitioning and regression trees (2019). https://CRAN.R-project.org/package=rpart. R package version 4.1-15
  46. 46.
    Urbanowicz, R.J., Meeker, M., La Cava, W., Olson, R.S., Moore, J.H.: Relief-based feature selection: introduction and review. J. Biomed. Inform. 85, 189–203 (2018).  https://doi.org/10.1016/j.jbi.2018.07.014. https://www.sciencedirect.com/science/article/pii/S1532046418301400CrossRefGoogle Scholar
  47. 47.
    Wu, Y., He, K.: Group normalization. Int. J. Comput. Vis. 128(3), 742–755 (2020).  https://doi.org/10.1007/s11263-019-01198-wCrossRefGoogle Scholar
  48. 48.
    Xu, L., Hutter, F., Hoos, H., Leyton-Brown, K.: Evaluating component solver contributions to portfolio-based algorithm selectors. In: Cimatti, A., Sebastiani, R. (eds.) SAT 2012. LNCS, vol. 7317, pp. 228–241. Springer, Heidelberg (2012).  https://doi.org/10.1007/978-3-642-31612-8_18CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Moritz Seiler
    • 1
    Email author
  • Janina Pohl
    • 1
  • Jakob Bossek
    • 2
  • Pascal Kerschke
    • 1
  • Heike Trautmann
    • 1
  1. 1.Statistics and Optimization GroupUniversity of MünsterMünsterGermany
  2. 2.Optimisation and LogisticsThe University of AdelaideAdelaideAustralia

Personalised recommendations