An Initialization Strategy for High-Dimensional Surrogate-Based Expensive Black-Box Optimization

Conference paper
Part of the Springer Proceedings in Mathematics & Statistics book series (PROMS, volume 62)

Abstract

Surrogate-based optimization methods build surrogate models of expensive black-box objective and constraint functions using previously evaluated points and use these models to guide the search for an optimal solution. These methods require considerably more computational overhead and memory than other optimization methods, so their applicability to high-dimensional problems is somewhat limited. Many surrogates, such as radial basis functions (RBFs) with linear polynomial tails, require a maximal set of affinely independent points to fit the initial model. This paper proposes an initialization strategy for surrogate-based methods called underdetermined simplex gradient descent (USGD) that uses underdetermined simplex gradients to make progress towards the optimum while building a maximal set of affinely independent points. Numerical experiments on a 72-dimensional groundwater bioremediation problem and on 200-dimensional and 1000-dimensional instances of 16 well-known test problems demonstrate that the proposed USGD initialization strategy yields dramatic improvements in the objective function value compared to standard initialization procedures. Moreover, USGD initialization substantially improves the performance of two optimization algorithms that use RBF surrogates compared to standard initialization methods on the same test problems.

Keywords

Engineering optimization High-dimensional black-box optimization Simplex gradient Surrogate model Function approximation Radial basis function Expensive function 

Notes

Acknowledgements

I would like to thank Ismael Vaz and Luís Vicente for the PSwarm package, that includes a pattern search Matlab code with options for combining with RBF models. I am also grateful to Jorge Moré and Stefan Wild for their Matlab code that creates performance and data profiles.

References

  1. 1.
    Abramson, M.A., Audet, C.: Convergence of mesh adaptive direct search to second-order stationary points. SIAM J. Optim. 17(2), 606–619 (2006)MathSciNetCrossRefMATHGoogle Scholar
  2. 2.
    Aleman, D.M., Romeijn, H.E., Dempsey, J.F.: A response surface approach to beam orientation optimization in intensity modulated radiation therapy treatment planning. INFORMS J. Comput. 21(1), 62–76 (2009)CrossRefGoogle Scholar
  3. 3.
    Audet, C., Dennis, J.E., Jr.: Mesh adaptive direct search algorithms for constrained optimization. SIAM J. Optim. 17(2), 188–217 (2006)MathSciNetCrossRefMATHGoogle Scholar
  4. 4.
    Audet, C., Dennis, J.E., Jr.: Le Digabel S.: Parallel space decomposition of the mesh adaptive direct search algorithm. SIAM J. Optim. 19(3), 1150–1170 (2008)Google Scholar
  5. 5.
    Bettonvil, B., Kleijnen J.P.C.: Searching for important factors in simulation models with many factors: Sequential bifurcation. Eur. J. Oper. Res. 96(1), 180–194 (1997)CrossRefMATHGoogle Scholar
  6. 6.
    Björkman, M., Holmström, K.: Global optimization of costly nonconvex functions using radial basis functions. Optim. Eng. 1(4), 373–397 (2000)MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    Booker, A.J., Dennis, J.E., Jr., Frank, P.D., Serafini, D.B., Torczon, V., Trosset M.W.: A rigorous framework for optimization of expensive functions by surrogates. Struct. Optim. 17(1), 1–13 (1999)CrossRefGoogle Scholar
  8. 8.
    Buhmann, M.D.: Radial Basis Functions. Cambridge University Press, Cambridge (2003)CrossRefMATHGoogle Scholar
  9. 9.
    Bull, A.D.: Convergence rates of efficient global optimization algorithms. J. Mach. Learn. Res. 12(Oct), 2879–2904 (2011)MathSciNetGoogle Scholar
  10. 10.
    Cassioli, A., Schoen, F.: Global optimization of expensive black box problems with a known lower bound. J. Global Optim. (2011). doi: 10.1007/s10898-011-9834-7Google Scholar
  11. 11.
    Chambers, M., Mount-Campbell, C.A.: Process optimization via neural network metamodeling. Int. J. Prod. Econ. 79(2), 93–100 (2002)CrossRefGoogle Scholar
  12. 12.
    Chen, L.-L., Liao, C., Lin, W.-B., Chang, L., Zhong, X.-M.: Hybrid-surrogate-model-based efficient global optimization for high-dimensional antenna design. Prog. Electromagnetics Res. 124, 85–100 (2012)CrossRefGoogle Scholar
  13. 13.
    Conn, A.R., Scheinberg, K., Vicente, L.N.: Geometry of interpolation sets in derivative free optimization. Math. Program. 111(1–2), 141–172 (2008a)MathSciNetMATHGoogle Scholar
  14. 14.
    Conn, A.R., Scheinberg, K., Vicente, L.N.: Geometry of sample sets in derivative-free optimization: polynomial regression and underdetermined interpolation. IMA J. Numer. Anal. 28(4), 721–748 (2008b)MathSciNetCrossRefMATHGoogle Scholar
  15. 15.
    Conn, A.R., Scheinberg, K., Vicente, L.N.: Global convergence of general derivative-free trust-region algorithms to first- and second-order critical points. SIAM J. Optim. 20(1), 387–415 (2009a)MathSciNetCrossRefMATHGoogle Scholar
  16. 16.
    Conn, A.R., Scheinberg, K., Vicente, L.N.: Introduction to Derivative-Free Optimization. SIAM, Philadelphia, PA (2009b)CrossRefMATHGoogle Scholar
  17. 17.
    Conn, A.R., Scheinberg, K., Toint, Ph.L.: Recent progress in unconstrained nonlinear optimization without derivatives. Math. Program. 79(3), 397–414 (1997)MathSciNetMATHGoogle Scholar
  18. 18.
    Conn, A.R., Le Digabel, S.: Use of quadratic models with mesh-adaptive direct search for constrained black box optimization. Optim. Meth. Software 28(1), 139–158 (2013)CrossRefMATHGoogle Scholar
  19. 19.
    Cressie, N.: Statistics for Spatial Data. Wiley, New York (1993)Google Scholar
  20. 20.
    Custódio, A.L., Rocha, H., Vicente, L.N.: Incorporating minimum Frobenius norm models in direct search. Computat. Optim. Appl. 46(2), 265–278 (2010)CrossRefMATHGoogle Scholar
  21. 21.
    Custódio, A.L., Vicente, L.N.: Using sampling and simplex derivatives in pattern search methods. SIAM J. Optim. 18(2), 537–555 (2007)MathSciNetCrossRefMATHGoogle Scholar
  22. 22.
    Egea, J.A., Vazquez, E., Banga, J.R., Marti, R.: Improved scatter search for the global optimization of computationally expensive dynamic models. J. Global Optim. 43(2–3), 175–190 (2009)CrossRefMATHGoogle Scholar
  23. 23.
    García-Palomares, U.M., García-Urrea, I.J.,  Rodríguez-Hernández, P.S.: On sequential and parallel non-monotone derivative-free algorithms for box constrained optimization. Optim. Meth. Software (2012). doi:10.1080/10556788.2012.693926Google Scholar
  24. 24.
    Gray, G.A., Kolda, T.G.: Algorithm 856: APPSPACK 4.0: asynchronous parallel pattern search for derivative-free optimization. ACM Trans. Math. Software 32(3), 485–507 (2006)Google Scholar
  25. 25.
    Gutmann, H.-M.: A radial basis function method for global optimization. J. Global Optim. 19(3), 201–227 (2001)MathSciNetCrossRefMATHGoogle Scholar
  26. 26.
    Hansen, N.: The CMA evolution strategy: a comparing review. In: Lozano, J.A., Larranga, P., Inza, I., Bengoetxea, E. (eds.) Towards a New Evolutionary Computation, pp. 75–102, Springer, Berlin (2006)Google Scholar
  27. 27.
    Hansen, N., Ostermeier, A.: Completely derandomized self-adaptation in evolution strategies. Evol. Comput. 9(2), 159–195 (2001)CrossRefGoogle Scholar
  28. 28.
    Holmström, K.: An adaptive radial basis algorithm (ARBF) for expensive black-box global optimization. J. Global Optim. 41(3), 447–464 (2008)MathSciNetCrossRefMATHGoogle Scholar
  29. 29.
    Huang, D., Allen, T.T., Notz, W.I., Zeng, N.: Global optimization of stochastic black-box systems via sequential kriging meta-models. J. Global Optim. 34(3), 441–466 (2006)MathSciNetCrossRefMATHGoogle Scholar
  30. 30.
    Jakobsson, S., Patriksson, M., Rudholm, J., Wojciechowski, A.: A method for simulation based optimization using radial basis functions. Optim. Eng. 11(4), 501–532 (2010)MathSciNetCrossRefMATHGoogle Scholar
  31. 31.
    Jin Y.: Surrogate-assisted evolutionary computation: recent advances and future challenges. Swarm Evol. Comput. 1(2), 61–70 (2011)CrossRefGoogle Scholar
  32. 32.
    Jin, Y., Olhofer, M., Sendhoff, B.: A framework for evolutionary optimization with approximate fitness functions. IEEE Trans. Evol. Comput. 6(5), 481–494 (2002)CrossRefGoogle Scholar
  33. 33.
    Jones, D.R., Schonlau, M., Welch, W.J.: Efficient global optimization of expensive black-box functions. J. Global Optim. 13(4), 455–492Google Scholar
  34. 34.
    Jones, D.R.: Large-scale multi-disciplinary mass optimization in the auto industry. Presented at the Modeling and Optimization: Theory and Applications (MOPTA) 2008 Conference, Ontario, Canada (2008)Google Scholar
  35. 35.
    Kolda, T.G., Lewis, R.M., Torczon, V.: Optimization by direct search: new perspectives on some classical and modern methods. SIAM Rev. 45(3), 385–482 (2003)MathSciNetCrossRefMATHGoogle Scholar
  36. 36.
    Kolda, T.G., Torczon, V.J.: On the convergence of asynchronous parallel pattern search. SIAM J. Optim. 14(4), 939–964 (2004)MathSciNetCrossRefMATHGoogle Scholar
  37. 37.
    Le Digabel, S.: Algorithm 909: NOMAD: Nonlinear optimization with the MADS algorithm. ACM Trans. Math. Software 37(4), 44:1–44:15 (2011)Google Scholar
  38. 38.
    Le Thi, H.A., Vaz, A.I.F., Vicente, L.N.: Optimizing radial basis functions by D.C. programming and its use in direct search for global derivative-free optimization. TOP 20(1), 190–214 (2012)Google Scholar
  39. 39.
    Loshchilov, I., Schoenauer, M., Sebag, M.: Self-adaptive surrogate-assisted covariance matrix adaptation evolution strategy. In: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2012), ACM Press, New York (2012)Google Scholar
  40. 40.
    Marsden, A.L., Wang, M., Dennis, J.E., Jr., Moin, P.: Optimal aeroacoustic shape design using the surrogate management framework. Optim. Eng. 5(2), 235–262 (2004)MathSciNetCrossRefMATHGoogle Scholar
  41. 41.
    Minsker, B.S., Shoemaker, C.A.: Dynamic optimal control of in-situ bioremediation of groundwater. J. Water Resour. Plann. Manag. 124(3), 149–161 (1998)CrossRefGoogle Scholar
  42. 42.
    Moré, J., Wild, S.: Benchmarking derivative-free optimization algorithms. SIAM J. Optim. 20(1), 172–191 (2009)MathSciNetCrossRefMATHGoogle Scholar
  43. 43.
    Moré, J., Garbow, B., Hillstrom, K.: Testing unconstrained optimization software. ACM Trans. Math. Software 7(1), 17–41 (1981)MathSciNetCrossRefMATHGoogle Scholar
  44. 44.
    Myers, R.H., Montgomery, D.C.: Response Surface Methodology: Process and Product Optimization Using Designed Experiments, 3rd edn. Wiley, New York (2009)Google Scholar
  45. 45.
    Oeuvray, R., Bierlaire, M.: BOOSTERS: A derivative-free algorithm based on radial basis functions. Int. J. Model. Simulat. 29(1), 26–36 (2009)Google Scholar
  46. 46.
    Oeuvray, R.: Trust-region methods based on radial basis functions with application to biomedical imaging. Ph.D. thesis, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland (2005)Google Scholar
  47. 47.
    Parr, J.M., Keane, A.J., Forrester, A.I.J., Holden, C.M.E.: Infill sampling criteria for surrogate-based optimization with constraint handling. Eng. Optim. 44(10), 1147–1166 (2012)CrossRefGoogle Scholar
  48. 48.
    Plantenga, T., Kolda, T.: HOPSPACK: Software framework for parallel derivative-free optimization. Sandia Technical Report (SAND 2009–6265). (2009)Google Scholar
  49. 49.
    Powell, M.J.D.: The theory of radial basis function approximation in 1990. In: Light, W. (ed.) Advances in Numerical Analysis, Volume 2: Wavelets, Subdivision Algorithms and Radial Basis Functions. pp. 105–210. Oxford University Press, Oxford (1992)Google Scholar
  50. 50.
    Powell, M.J.D.: UOBYQA: Unconstrained optimization by quadratic approximation. Math. Program. 92(3), 555–582 (2002)MathSciNetCrossRefMATHGoogle Scholar
  51. 51.
    Powell, M.J.D.: The NEWUOA software for unconstrained optimization without derivatives. In: Di Pillo, G., Roma, M. (eds.) Large-Scale Nonlinear Optimization, pp. 255–297. Springer, New York (2006)CrossRefGoogle Scholar
  52. 52.
    Regis, R.G.: Stochastic radial basis function algorithms for large-scale optimization involving expensive black-box objective and constraint functions. Comput. Oper. Res. 38(5), 837–853 (2011)MathSciNetCrossRefGoogle Scholar
  53. 53.
    Regis, R.G.: Constrained optimization by radial basis function interpolation for high-dimensional expensive black-box problems with infeasible initial points. Eng. Optim. (2013). doi: 10.1080/0305215X.2013.765000.MathSciNetGoogle Scholar
  54. 54.
    Regis, R.G., Shoemaker, C.A.: A stochastic radial basis function method for the global optimization of expensive functions. INFORMS J. Comput. 19(4), 497–509 (2007a)MathSciNetCrossRefMATHGoogle Scholar
  55. 55.
    Regis, R.G., Shoemaker, C.A.: Improved strategies for radial basis function methods for global optimization. J. Global Optim. 37(1), 113–135 (2007b)MathSciNetCrossRefMATHGoogle Scholar
  56. 56.
    Regis, R.G., Shoemaker, C.A.: A quasi-multistart framework for global optimization of expensive functions using response surface models. J. Global Optim. (2012). doi: 10.1007/s10898-012-9940-1Google Scholar
  57. 57.
    Regis, R.G., Shoemaker, C.A.: Combining radial basis function surrogates and dynamic coordinate search in high-dimensional expensive black-box optimization. Eng. Optim. 45(5), 529–555 (2013)MathSciNetCrossRefGoogle Scholar
  58. 58.
    Rocha, H., Dias, J.M., Ferreira, B.C., Lopes, M.C.: Selection of intensity modulated radiation therapy treatment beam directions using radial basis functions within a pattern search methods framework. J. Global Optim. (2012). doi: 10.1007/s10898-012-0002-5Google Scholar
  59. 59.
    Sacks, J., Welch, W.J., Mitchell, T.J., Wynn, H.P.: Design and analysis of computer experiments. Stat. Sci. 4(4), 409–435 (1989)MathSciNetCrossRefMATHGoogle Scholar
  60. 60.
    Scheinberg, K., Toint, Ph.L.: Self-correcting geometry in model-based algorithms for derivative-free unconstrained optimization. SIAM J. Optim. 20(6), 3512–3532 (2010)MathSciNetCrossRefMATHGoogle Scholar
  61. 61.
    Shan, S., Wang, G.: Survey of modeling and optimization strategies to solve high-dimensional design problems with computationally-expensive black-box functions. Struct. Multidisci- plinary Optim. 41(2), 219–241 (2010)MathSciNetCrossRefMATHGoogle Scholar
  62. 62.
    Shan, S., Wang, G.G.: Metamodeling for high dimensional simulation-based design problems. ASME Journal of Mechanical Design, 132(5), 051009 (2011)MathSciNetCrossRefGoogle Scholar
  63. 63.
    Shoemaker, C.A., Willis, M., Zhang, W., Gossett, J.: Model analysis of reductive dechlorination with data from Cape Canaveral field site. In: Magar, V., Vogel, T., Aelion, C., Leeson, A. (eds.) Innovative Methods in Support of Bioremediation, pp. 125–131. Battelle Press, Columbus, OH (2001)Google Scholar
  64. 64.
    Tolson, B.A., Shoemaker, C.A.: Dynamically dimensioned search algorithm for computationally efficient watershed model calibration. Water Resour. Res. 43, W01413 (2007) doi:10.1029/2005WR004723.CrossRefGoogle Scholar
  65. 65.
    Torczon, V.: On the convergence of pattern search algorithms. SIAM J. Optim. 7(1), 1–25 (1997)MathSciNetCrossRefMATHGoogle Scholar
  66. 66.
    Vapnik, V.: The Nature of Statistical Learning Theory. Springer, New York (1995)CrossRefMATHGoogle Scholar
  67. 67.
    Vaz, A.I.F., Vicente, L.N.: A particle swarm pattern search method for bound constrained global optimization. J. Global Optim. 39(2), 197–219 (2007)MathSciNetCrossRefMATHGoogle Scholar
  68. 68.
    Vaz, A.I.F., Vicente, L.N.: PSwarm: A hybrid solver for linearly constrained global derivative-free optimization. Optim. Meth. Software 24(4–5), 669–685 (2009)MathSciNetCrossRefMATHGoogle Scholar
  69. 69.
    Vazquez, E., Bect, J.: Convergence properties of the expected improvement algorithm with fixed mean and covariance functions. J. Stat. Plann. Infer. 140(11), 3088–3095 (2010)MathSciNetCrossRefMATHGoogle Scholar
  70. 70.
    Viana, F.A.C., Haftka, R.T., Watson, L.T.: Why not run the efficient global optimization algorithm with multiple surrogates?. 51th AIAA/ASME /ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, AIAA 2010–3090, Orlando (2010)Google Scholar
  71. 71.
    Villemonteix, J., Vazquez, E., Walter, E.: An informational approach to the global optimization of expensive-to-evaluate functions. J. Global Optim. 44(4), 509–534 (2009)MathSciNetCrossRefMATHGoogle Scholar
  72. 72.
    Wild, S.M., Regis, R.G., Shoemaker, C.A.: ORBIT: Optimization by radial basis function interpolation in trust-regions. SIAM J. Sci. Comput. 30(6), 3197–3219 (2008)MathSciNetCrossRefMATHGoogle Scholar
  73. 73.
    Wild, S.M., Shoemaker, C.A.: Global convergence of radial basis function trust region derivative-free algorithms. SIAM J. Optim. 21(3), 761–781 (2011)MathSciNetCrossRefMATHGoogle Scholar
  74. 74.
    Yoon, J.-H., Shoemaker, C.A.: Comparison of optimization methods for ground-water bioremediation. J. Water Resour. Plann. Manag. 125(1), 54–63 (1999)CrossRefGoogle Scholar
  75. 75.
    The MathWorks, Inc. Matlab Optimization Toolbox: User’s Guide, Version 4. Natick, MA (2009)Google Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  1. 1.Department of MathematicsSaint Joseph’s UniversityPhiladelphiaUSA

Personalised recommendations