Skip to main content
SpringerLink
Log in

Learning a Mixture of Search Heuristics

  • Chapter
Autonomous Search

Abstract

This chapter describes programs that solve constraint satisfaction problems with multiple heuristics. It demonstrates the varied efficacy of individual constraint-solving metrics and the potential power available from a mixture of heuristics that references them. It describes a weighted-mixture decision process, and explains how one autonomous learner constructs its own labeled training examples from its search experience, and then learns a weighted mixture from them. Four new techniques are introduced to manage a large body of conflicting heuristics, and illustrated with empirical results.

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 84.99
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 109.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info
Hardcover Book
USD 109.99
Price excludes VAT (USA)
  • Durable hardcover 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

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Aardal, K. I., Hoesel, S. P. M. v., Koster, A. M. C. A., Mannino, C., Sassano, A.: Models and solution techniques for frequency assignment problems. A Quarterly Journal of Operations Research, 1(4), pp. 261–317 (2003)

    MATH  Google Scholar 

  2. Ali, K., Pazzani, M.: Error reduction through learning multiple descriptions. Machine Learning, 24, pp. 173–202 (1996)

    Google Scholar 

  3. Arbelaez, A., Hamadi, Y., Sebag, M.: Online heuristic selection in constraint programming. International Symposium on Combinatorial Search (SoCS-2009), Lake Arrowhead, CA

    Google Scholar 

  4. Birattari, M., Stützle, T., Paquete, L., Varrentrapp, K.: (2002). A Racing Algorithm for configuring metaheuristics. Genetic and Evolutionary Computation Conference (GECCO-2002), New York, NY, pp. 11–18 (2009)

    Google Scholar 

  5. Borrett, J. E., Tsang, E. P. K., Walsh, N. R.: Adaptive constraint satisfaction: The quickest first principle. 12th European Conference on AI (ECAI-1996), Budapest, Hungary, pp. 160–164 (1996)

    Google Scholar 

  6. Boussemart, F., Hemery, F., Lecoutre, C., Sais, L.: Boosting systematic search by weighting constraints. 16th European Conference on AI (ECAI-2004), Valencia, Spain, pp. 146–150 (2004)

    Google Scholar 

  7. Burke, E., Hart, E., Kendall, G., Newall, J., Ross, P., Schulenburg, S.: Hyperheuristics: an emerging direction in modern research technology. Handbook of Metaheuristics, Dordrecht: Kluwer Academic Publishers, pp. 457–474 (2003)

    Google Scholar 

  8. Chakhlevitch, K., Cowling, P.: Hyperheuristics: Recent developments. C. Cotta, M. Sevaux, K. Sorensen (Eds.), Adaptive and Multilevel Metaheuristics, Studies in Computational Intelligence, Springer, pp. 3–29 (2008)

    Chapter  Google Scholar 

  9. Cicirello, V. A., Smith, S. F.: Heuristic selection for stochastic search optimization: Modeling solution quality by extreme value theory. Principles and Practice of Constraint Programming (CP-2004), Toronto, Canada, LNCS 3258, pp. 197–211 (2004)

    Chapter  Google Scholar 

  10. D’Andrade, R. G.: Some propositions about the relations between culture and human cognition. J. W. Stigler, R. A. Shweder, G. Herdt (Eds.), Cultural Psychology: Essays on Comparative Human Development, Cambridge University Press, Cambridge 1990, pp. 65–129 (1990)

    Google Scholar 

  11. Dietterich, T. G.: Ensemble methods in machine learning. J. Kittler, F. Roli (Eds.) First International Workshop on Multiple Classifier Systems, MCS-2000, Cagliari, Italy, LNCS 1857, pp. 1–15 (2000)

    Google Scholar 

  12. Epstein, S. L.: For the Right Reasons: The FORR architecture for learning in a skill domain. Cognitive Science, 18, 479–511 (1994)

    Article  Google Scholar 

  13. Epstein, S. L., Freuder, E. C., Wallace, R.: Learning to support constraint programmers. Computational Intelligence, 21(4), 337–371 (2005)

    Article  MathSciNet  Google Scholar 

  14. Freund, Y., Schapire, R.: Experiments with a new boosting algorithm. Thirteenth International Conference on Machine Learning (ICML-96), Bari, Italy, pp. 148–156 (1996)

    Google Scholar 

  15. Fukunaga, A. S.: Automated discovery of composite SAT variable-selection heuristics. Eighteenth National Conference on Artificial Intelligence (AAAI-2002), Edmonton, Canada, pp. 641–648 (2002)

    Google Scholar 

  16. Gagliolo, M., Schmidhuber, J.: Dynamic algorithm portfolios. Ninth International Symposium on Artificial Intelligence and Mathematics, Special Issue of the Annals of Mathematics and Artificial Intelligence, 47(3-4), pp. 295–328 (2006)

    Article  MATH  MathSciNet  Google Scholar 

  17. Gebruers, C., Hnich, B., Bridge, D., Freuder, E.: Using CBR to select solution strategies in constraint programming. Case-Based Reasoning Research and Development. LNCS 3620, pp. 222–236 (2005)

    Chapter  Google Scholar 

  18. Gent, I. P., Prosser, P., Walsh, T.: The constrainedness of search. The Thirteenth National Conference on Artificial Intelligence (AAAI-1996), Portland, Oregon, pp. 246–252 (1996)

    Google Scholar 

  19. Gomes, C., Fernandez, C., Selman, B., Bessière, C.: Statistical regimes across constrainedness regions. Principles and Practice of Constraint Programming (CP-2004), Toronto, Canada, LNCS 3258, pp. 32–46 (2004)

    Chapter  Google Scholar 

  20. Gomes, C. P., Selman, B., Kautz, H. A.: Boosting combinatorial search through randomization, The Fifteenth National Conference on Artificial Intelligence (AAAI-1998), Madison, Wisconsin, pp. 431–437 (1998)

    Google Scholar 

  21. Gomes, C. P., Selman, B.: Algorithm portfolios. Artificial Intelligence, 126(1-2), pp. 43–62 (2001)

    Article  MATH  MathSciNet  Google Scholar 

  22. Hansen, L., Salamon, P.: Neural network ensembles. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12, pp. 993–1001 (1990)

    Article  Google Scholar 

  23. Hulubei, T., O’Sullivan, B.: Search heuristics and heavy-tailed behavior. Principles and Practice of Constraint Programming (CP-2005), Barcelona, Spain, pp. 328–342 (2005)

    Chapter  Google Scholar 

  24. Hutter, F., Hamadi, Y., Hoos, H. H., Leyton-Brown, K.: Performance prediction and automated tuning of randomized and parametric algorithms. Principles and Practice of Constraint Programming (CP-2006), Nantes, France, pp. 213–228 (2006)

    Chapter  Google Scholar 

  25. Johnson, D. S., Aragon, C. R., McGeoch, L. A., Schevon, C.: Optimization by simulated annealing: An experimental evaluation; Part I, Graph Partitioning. Operations Research, 37(6), 865–892 (1989)

    Article  MATH  Google Scholar 

  26. Kivinen, J., Warmuth, M. K.: Averaging expert predictions. Computational Learning Theory: 4th European Conference (EuroCOLT-1999), Nordkirchen, Germany, pp. 153–167 (1999)

    Google Scholar 

  27. Lecoutre, C., Boussemart, F., Hemery, F.: Backjump-based techniques versus conflict directed heuristics. 16th IEEE International Conference on Tools with Artificial Intelligence (ICTAI-2004), Boca Raton, Florida, USA, pp. 549–557 (2004)

    Chapter  Google Scholar 

  28. Minton, S., Allen, J. A., Wolfe, S., Philpot, A.: An overview of learning in the Multi-TAC system. First International Joint Workshop on Artificial Intelligence and Operations Research, Timberline, Oregon, USA (1995)

    Google Scholar 

  29. Nareyek, A.: Choosing search heuristics by non-stationary reinforcement learning. Metaheuristics: Computer Decision-Making, Kluwer Academic Publishers, Norwell, MA, USA, pp. 523–544 (2004)

    Google Scholar 

  30. O’Mahony, E., Hebrard, E., Holland, A., Nugent, C., O’Sullivan, B.: Using case-based reasoning in an algorithm portfolio for constraint solving. 19th Irish Conference on Artificial Intelligence and Cognitive Science (AICS-2008), Cork, Ireland (2008)

    Google Scholar 

  31. Opitz, D., Shavlik, J.: Generating accurate and diverse members of a neural-network ensemble. Advances in Neural Information Processing Systems, 8, pp. 535–541 (1996)

    Google Scholar 

  32. Otten, L., Gronkvist, M., Dubhashi, D. P.: Randomization in constraint programming for airline planning. Principles and Practice of Constraint Programming (CP-2006), Nantes, France, pp. 406–420 (2006)

    Chapter  Google Scholar 

  33. Petrie, K. E., Smith, B. M.: Symmetry breaking in graceful graphs. Principles and Practice of Constraint Programming (CP-2003), Kinsale, Ireland, LNCS 2833, pp. 930–934 (2003)

    Chapter  Google Scholar 

  34. Petrovic, S., Epstein, S. L.: Full restart speeds learning. 19th International FLAIRS Conference (FLAIRS-06), Melbourne Beach, Florida (2006)

    Google Scholar 

  35. Petrovic, S., Epstein, S. L.: Relative support weight learning for constraint solving. Workshop on Learning for Search, AAAI-06, Boston, Massachusetts, USA, pp. 115–122 (2006)

    Google Scholar 

  36. Petrovic, S., Epstein, S. L.: Preferences improve learning to solve constraint problems. Workshop on Preference Handling for Artificial Intelligence (AAAI-07), Vancouver, Canada, pp. 71–78 (2007)

    Google Scholar 

  37. Petrovic, S., Epstein, S. L.: Random subsets support learning a mixture of heuristics. International Journal on Artificial Intelligence Tools (IJAIT), 17(3), pp. 501–520 (2008)

    Article  Google Scholar 

  38. Ruan, Y., Kautz, H., Horvitz, E.: The backdoor key: A path to understanding problem hardness. Nineteenth National Conference on Artificial Intelligence (AAAI-2004), San Jose, CA, USA, pp. 124–130 (2004)

    Google Scholar 

  39. Saari, D. G.: Geometry of voting. Studies in Economic Theory, Vol. 3, Springer (1994)

    MATH  Google Scholar 

  40. Sabin, D., Freuder, E. C.: Understanding and improving the MAC algorithm. Principles and Practice of Constraint Programming (CP-1997), Linz, Austria, LNCS 1330, pp. 167–181 (1997)

    Google Scholar 

  41. Samulowitz, H., Memisevic, R.: Learning to solve QBF. 22nd National Conference on Artificial intelligence (AAAI-2007), Vancouver, Canada, pp. 255–260 (2007)

    Google Scholar 

  42. Schapire, R. E.: The strength of weak learnability. Machine Learning, 5(2): 197–227 (1990)

    Google Scholar 

  43. Smith, B., Grant, S.: Trying harder to fail first. European Conference on Artificial Intelligence (ECAI-1998), pp. 249–253 (1998)

    Google Scholar 

  44. Streeter, M., Golovin, D., Smith, S. F.: Combining multiple heuristics online. 22nd National Conference on Artificial Intelligence (AAAI-07), Vancouver, Canada, pp. 1197–1203 (2007)

    Google Scholar 

  45. Terashima-Marin, H., Ortiz-Bayliss, J., Ross, P., Valenzuela-Rendón, M.: Using hyper-heuristics for the dynamic variable ordering in binary Constraint Satisfaction Problems, MICAI 2008, Advances in Artificial Intelligence, LNCS 5317, pp. 407–417 (2008)

    Google Scholar 

  46. Valentini, G., Masulli, F.: Ensembles of learning machines. Neural Nets (WIRN-2002), Vietri sul Mare, Italy, LNCS 2486, pp. 3–22 (2002)

    Chapter  Google Scholar 

  47. Wallace, R. J.: Factor analytic studies of CSP heuristics. Principles and Practice of Constraint Programming (CP-2005), Barcelona, Spain, LNCS 3709, pp. 712–726 (2005)

    Chapter  Google Scholar 

  48. Whitley, D.: The genitor algorithm and selection pressure: Why rank-based allocation of reproductive trials is best. International Conference on Genetic Algorithms (ICGA-1989), pp. 116–121 (1989)

    Google Scholar 

  49. Xu, L., Hutter, F., Hoos, H., Leyton-Brown, K.: SATzilla-07: The design and analysis of an algorithm portfolio for SAT. Principles and Practice of Constraint Programming (CP-2007), Providence, RI, USA, pp. 712–727 (2007)

    Chapter  Google Scholar 

  50. Young, H. P.: Condorcet’s theory of voting. The American Political Science Review, 82(4), 1231–1244 (1988)

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science, Hunter College and The Graduate Center of The City University of New York, New York, New York, USA

    Susan L. Epstein

Authors
  1. Susan L. Epstein
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Smiljana Petrovic
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Susan L. Epstein .

Editor information

Editors and Affiliations

  1. Microsoft Research Cambridge, JJ Thomson Avenue 7, Cambridge, CB3 0FB, United Kingdom

    Youssef Hamadi

  2. Federico Santa María, Departamento de Informática, Universidad Técnica, Avenida España 1680, Valparaíso, Chile

    Eric Monfroy

  3. Faculté des Sciences, LERIA, Université d'Angers, Boulevard Lavoisier 2, Angers CX 01, 49045, France

    Frédéric Saubion

Rights and permissions

Reprints and permissions

Copyright information

© 2011 Springer-Verlag Berlin Heidelberg

About this chapter

Cite this chapter

Epstein, S.L., Petrovic, S. (2011). Learning a Mixture of Search Heuristics. In: Hamadi, Y., Monfroy, E., Saubion, F. (eds) Autonomous Search. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-21434-9_5

Download citation

  • DOI: https://doi.org/10.1007/978-3-642-21434-9_5

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-642-21433-2

  • Online ISBN: 978-3-642-21434-9

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation