Skip to main content
SpringerLink
Log in

Model-Based Algorithm Configuration with Default-Guided Probabilistic Sampling

  • Conference paper
  • First Online:
Parallel Problem Solving from Nature – PPSN XVI (PPSN 2020)

Part of the book series: Lecture Notes in Computer Science ((LNTCS,volume 12269))

Included in the following conference series:

Abstract

In recent years, general-purpose automated algorithm configuration procedures have enabled impressive improvements in the state of the art in solving a wide range of challenging problems from AI, operations research and other areas. To search vast combinatorial spaces of parameter settings for a given algorithm as efficiently as possible, the most successful configurators combine techniques such as racing, estimation of distribution algorithms, Bayesian optimisation and model-free stochastic search. Two of the most widely used general-purpose algorithm configurators, SMAC and irace, can be seen as combinations of Bayesian optimisation and racing, and of racing and an estimation of distribution algorithm, respectively. Here, we propose a first approach that combines all three of these techniques into one single configurator, while exploiting prior knowledge contained in expert-chosen default parameter values. We demonstrate significant performance improvements over irace and SMAC on a broad range of running time optimisation scenarios from AClib.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 79.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 99.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

References

  1. Ahmadizadeh, K., Dilkina, B., Gomes, C.P., Sabharwal, A.: An empirical study of optimization for maximizing diffusion in networks. In: Cohen, D. (ed.) CP 2010. LNCS, vol. 6308, pp. 514–521. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-15396-9_41

    Chapter  Google Scholar 

  2. Anastacio, M., Luo, C., Hoos, H.: Exploitation of default parameter values in automated algorithm configuration. In: Workshop Data Science Meets Optimisation (DSO), IJCAI 2019, August 2019

    Google Scholar 

  3. Ansótegui, C., Malitsky, Y., Samulowitz, H., Sellmann, M., Tierney, K.: Model-based genetic algorithms for algorithm configuration. In: Proceedings of the IJCAI 2015, pp. 733–739 (2015)

    Google Scholar 

  4. Ansótegui, C., Sellmann, M., Tierney, K.: A gender-based genetic algorithm for the automatic configuration of algorithms. In: Gent, I.P. (ed.) CP 2009. LNCS, vol. 5732, pp. 142–157. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-04244-7_14

    Chapter  Google Scholar 

  5. Atamtürk, A., Muñoz, J.C.: A study of the lot-sizing polytope. Math. Program. 99(3), 443–465 (2004). https://doi.org/10.1007/s10107-003-0465-8

    Article  MathSciNet  MATH  Google Scholar 

  6. Balint, A., Manthey, N.: SparrowToRiss. In: Proceedings of the SAT Competition 2014, pp. 77–78 (2014)

    Google Scholar 

  7. Biere, A.: Yet another local search solver and lingeling and friends entering the SAT competition 2014 (2014)

    Google Scholar 

  8. Birattari, M., Yuan, Z., Balaprakash, P., Stützle, T.: F-Race and iterated F-Race: an overview. In: Bartz-Beielstein, T., Chiarandini, M., Paquete, L., Preuss, M. (eds.) Experimental Methods for the Analysis of Optimization Algorithms, pp. 311–336. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-02538-9_13

    Chapter  Google Scholar 

  9. Brummayer, R., Lonsing, F., Biere, A.: Automated testing and debugging of SAT and QBF solvers. In: Strichman, O., Szeider, S. (eds.) SAT 2010. LNCS, vol. 6175, pp. 44–57. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-14186-7_6

    Chapter  MATH  Google Scholar 

  10. Cáceres, L.P., López-Ibáñez, M., Hoos, H., Stützle, T.: An experimental study of adaptive capping in irace. In: Battiti, R., Kvasov, D.E., Sergeyev, Y.D. (eds.) LION 2017. LNCS, vol. 10556, pp. 235–250. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-69404-7_17

    Chapter  Google Scholar 

  11. Domhan, T., Springenberg, J.T., Hutter, F.: Speeding up automatic hyperparameter optimization of deep neural networks by extrapolation of learning curves. In: Yang, Q., Wooldridge, M.J. (eds.) IJCAI 2015, pp. 3460–3468. AAAI Press (2015)

    Google Scholar 

  12. Fawcett, C., Helmert, M., Hoos, H., Karpas, E., Röger, G., Seipp, J.: FD-Autotune: domain-specific configuration using fast downward. In: Proceedings of the ICAPS Workshop, PAL 2011, pp. 13–20 (2011)

    Google Scholar 

  13. Fawcett, C., Hoos, H.H.: Analysing differences between algorithm configurations through ablation. J. Heuristics 22(4), 431–458 (2015). https://doi.org/10.1007/s10732-014-9275-9

    Article  Google Scholar 

  14. Gebser, M., Kaufmann, B., Schaub, T.: Conflict-driven answer set solving: from theory to practice. Artif. Intell. 187–188, 52–89 (2012)

    Article  MathSciNet  Google Scholar 

  15. Gerevini, A., Saetti, A., Serina, I.: Planning through stochastic local search and temporal action graphs in LPG. J. Artif. Intell. Res. 20, 239–290 (2003)

    Article  Google Scholar 

  16. Gerevini, A., Saetti, A., Serina, I.: An approach to efficient planning with numerical fluents and multi-criteria plan quality. Artif. Intell. 172, 899–944 (2008)

    Article  MathSciNet  Google Scholar 

  17. Gerevini, A., Saetti, A., Serina, I.: An empirical analysis of some heuristic features for planning through local search and action graphs. Fundamenta Informaticae 107(2–3), 167–197 (2011)

    Article  MathSciNet  Google Scholar 

  18. Gomes, C.P., van Hoeve, W.-J., Sabharwal, A.: Connections in networks: a hybrid approach. In: Perron, L., Trick, M.A. (eds.) CPAIOR 2008. LNCS, vol. 5015, pp. 303–307. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-68155-7_27

    Chapter  Google Scholar 

  19. Hauschild, M., Pelikan, M.: An introduction and survey of estimation of distribution algorithms. Swarm Evol. Comput. 1(3), 111–128 (2011)

    Article  Google Scholar 

  20. Hoos, H.H.: Automated algorithm configuration and parameter tuning. In: Hamadi, Y., Monfroy, E., Saubion, F. (eds.) Autonomous Search, pp. 37–71. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-21434-9_3

    Chapter  Google Scholar 

  21. Hoos, H.H.: Programming by optimization. Commun. ACM 55(2), 70–80 (2012)

    Article  Google Scholar 

  22. Hutter, F., Hoos, H.H., Leyton-Brown, K.: Automated configuration of mixed integer programming solvers. In: Lodi, A., Milano, M., Toth, P. (eds.) CPAIOR 2010. LNCS, vol. 6140, pp. 186–202. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-13520-0_23

    Chapter  Google Scholar 

  23. Hutter, F., Hoos, H.H., Leyton-Brown, K.: Sequential model-based optimization for general algorithm configuration. In: Coello, C.A.C. (ed.) LION 2011. LNCS, vol. 6683, pp. 507–523. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-25566-3_40

    Chapter  Google Scholar 

  24. Hutter, F., Hoos, H.H., Leyton-Brown, K., Stützle, T.: ParamILS: an automatic algorithm configuration framework. J. Artif. Intell. Res. 36, 267–306 (2009)

    Article  Google Scholar 

  25. Hutter, F., Lindauer, M., Balint, A., Bayless, S., Hoos, H.H., Leyton-Brown, K.: The configurable SAT solver challenge (CSSC). Artif. Intell. 243, 1–25 (2017)

    Article  MathSciNet  Google Scholar 

  26. Hutter, F., et al.: AClib: a benchmark library for algorithm configuration. In: Pardalos, P.M., Resende, M.G.C., Vogiatzis, C., Walteros, J.L. (eds.) LION 2014. LNCS, vol. 8426, pp. 36–40. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-09584-4_4

    Chapter  Google Scholar 

  27. Kotthoff, L., Thornton, C., Hoos, H.H., Hutter, F., Leyton-Brown, K.: Auto-WEKA 2.0: automatic model selection and hyperparameter optimization in WEKA. J. Mach. Learn. Res. 18, 25:1–25:5 (2017)

    MathSciNet  Google Scholar 

  28. Leyton-Brown, K., Pearson, M., Shoham, Y.: Towards a universal test suite for combinatorial auction algorithms. In: Jhingran, A., Mackie-Mason, J., Tygar, D.J. (eds.) Proceedings of the 2nd ACM Conference on Electronic Commerce (EC-00), Minneapolis, MN, USA, 17–20 October 2000, pp. 66–76. ACM (2000)

    Google Scholar 

  29. Lindauer, M., Hutter, F.: Warmstarting of model-based algorithm configuration. In: Proceedings of the AAAI-18, IAAI-18, and EAAI-18, pp. 1355–1362 (2018)

    Google Scholar 

  30. Long, D., Fox, M.: The 3rd international planning competition: results and analysis. J. Artif. Intell. Res. 20, 1–59 (2003)

    Article  Google Scholar 

  31. López-Ibáñez, M., Dubois-Lacoste, J., Pérez Cáceres, L., Stützle, T., Birattari, M.: The irace package: iterated racing for automatic algorithm configuration. Oper. Res. Perspect. 3, 43–58 (2016)

    Article  MathSciNet  Google Scholar 

  32. Luo, C., Hoos, H.H., Cai, S., Lin, Q., Zhang, H., Zhang, D.: Local search with efficient automatic configuration for minimum vertex cover. In: IJCAI-19, pp. 1297–1304. International Joint Conferences on Artificial Intelligence Organization (2019)

    Google Scholar 

  33. Mugrauer, F., Balint, A.: Sat encoded low autocorrelation binary sequence (labs) benchmark description. In: Proceedings of the SAT Competition 2013, pp. 117–118 (2013)

    Google Scholar 

  34. Penberthy, J.S., Weld, D.S.: Temporal planning with continuous change. In: Proceedings of the AAAI-94, vol. 2, pp. 1010–1015 (1994)

    Google Scholar 

  35. Pushak, Y., Hoos, H.: Algorithm configuration landscapes: more benign than expected? In: Auger, A., Fonseca, C.M., Lourenço, N., Machado, P., Paquete, L., Whitley, D. (eds.) PPSN 2018. LNCS, vol. 11102, pp. 271–283. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-99259-4_22

    Chapter  Google Scholar 

  36. Thornton, C., Hutter, F., Hoos, H.H., Leyton-Brown, K.: Auto-WEKA: combined selection and hyperparameter optimization of classification algorithms. In: The 19th ACM SIGKDD, KDD 2013, pp. 847–855 (2013)

    Google Scholar 

  37. Vallati, M., Fawcett, C., Gerevini, A., Hoos, H.H., Saetti, A.: Automatic generation of efficient domain-optimized planners from generic parametrized planners. In: Proceedings of the RCRA 2011, pp. 111–123 (2011)

    Google Scholar 

  38. Xu, L., Hutter, F., Hoos, H.H., Leyton-Brown, K.: Hydra-MIP: automated algorithm configuration and selection for mixed integer programming. In: Proceedings of the RCRA 2011, pp. 16–30 (2011)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Leiden University, Leiden, The Netherlands

    Marie Anastacio & Holger Hoos

  2. University of British Columbia, Vancouver, Canada

    Holger Hoos

Authors
  1. Marie Anastacio
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Holger Hoos
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Marie Anastacio .

Editor information

Editors and Affiliations

  1. Leiden University, Leiden, The Netherlands

    Thomas Bäck

  2. Leiden University, Leiden, The Netherlands

    Mike Preuss

  3. Leiden University, Leiden, The Netherlands

    André Deutz

  4. Sorbonne University, Paris, France

    Hao Wang

  5. Sorbonne University, Paris, France

    Carola Doerr

  6. Leiden University, Leiden, The Netherlands

    Michael Emmerich

  7. University of Münster, Münster, Germany

    Heike Trautmann

Appendix: Robustness to Misleading Default Values

Appendix: Robustness to Misleading Default Values

To obtain better insights into the robustness of our approach to misleading default values, we generated random configurations until we found one that performed worse than the default, but produced time-outs on fewer than a third of the training instances from our three SAT benchmarks. Then, we repeated a few configuration experiments from Sect. 5.1, using these new, misleading default configurations (using the same protocol as described earlier). The results of this experiment are shown in Table 5.

Table 5. Results for SMAC and SMAC+PS when given a default generated to be misleading. The numbers shown are median PAR10 scores in CPU seconds; best results are underlined, while boldface indicates results that are statistically tied to the best, according to a one-sided Mann-Whitney test (\(\alpha = 0.05\)).
Full size table

Rights and permissions

Reprints and permissions

Copyright information

© 2020 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Anastacio, M., Hoos, H. (2020). Model-Based Algorithm Configuration with Default-Guided Probabilistic Sampling. In: Bäck, T., et al. Parallel Problem Solving from Nature – PPSN XVI. PPSN 2020. Lecture Notes in Computer Science(), vol 12269. Springer, Cham. https://doi.org/10.1007/978-3-030-58112-1_7

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-58112-1_7

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-58111-4

  • Online ISBN: 978-3-030-58112-1

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation