Abstract
The Expected Hypervolume Improvement (EHVI) is a frequently used infill criterion in surrogate-assisted multi-criterion optimization. It needs to be frequently called during the execution of such algorithms. Despite recent advances in improving computational efficiency, its running time for three or more objectives has remained in \(O(n^d)\) for \(d\ge 3\), where d is the number of objective functions and n is the size of the incumbent Pareto-front approximation. This paper proposes a new integration scheme, which makes it possible to compute the EHVI in \(\varTheta (n \log n)\) optimal time for the important three-objective case (\(d=3\)). The new scheme allows for a generalization to higher dimensions and for computing the Probability of Improvement (PoI) integral efficiently. It is shown, both theoretically and empirically, that the hidden constant in the asymptotic notation is small. Empirical speed comparisons were designed between the C++ implementations of the new algorithm (which will be in the public domain) and those recently published by competitors, on randomly-generated non-dominated fronts of size 10, 100, and 1000. The experiments include the analysis of batch computations, in which only the parameters of the probability distribution change but the incumbent Pareto-front approximation stays the same. Experimental results show that the new algorithm is always faster than the other algorithms, sometimes over \(10^4\) times faster.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Also called Efficient Global Optimization [9].
References
Zaefferer, M., Bartz-Beielstein, T., Naujoks, B., Wagner, T., Emmerich, M.: A case study on multi-criteria optimization of an event detection software under limited budgets. In: Purshouse, R.C., Fleming, P.J., Fonseca, C.M., Greco, S., Shaw, J. (eds.) EMO 2013. LNCS, vol. 7811, pp. 756–770. Springer, Heidelberg (2013). doi:10.1007/978-3-642-37140-0_56
Yang, K., Deutz, A., Yang, Z., Bäck, T., Emmerich, M.: Truncated expected hypervolume improvement: exact computation and application. In: IEEE Congress on Evolutionary Computation (CEC). IEEE (2016)
Yang K, Gaida D, Bäck T, Emmerich M.: Expected hypervolume improvement algorithm for PID controller tuning and the multiobjective dynamical control of a biogas plant. In: 2015 IEEE Congress on Evolutionary Computation (CEC), pp. 1934–1942, May 2015
Michael, T.M., Giannakoglou, K.C., Naujoks, B.: Single-and multiobjective evolutionary optimization assisted by Gaussian random field metamodels. IEEE Trans. Evol. Comput. 10(4), 421–439 (2006)
Koch, P., Wagner, T., Emmerich, M.T., Bäck, T., Konen, W.: Efficient multi-criteria optimization on noisy machine learning problems. Appl. Soft Comput. 29, 357–370 (2015)
Shimoyama, K., Jeong, S., Obayashi, S.: Kriging-surrogate-based optimization considering expected hypervolume improvement in non-constrained many-objective test problems. In: IEEE Congress on Evolutionary Computation (CEC), pp. 658–665. IEEE (2013)
Shimoyama, K., Sato, K., Jeong, S., Obayashi, S.: Comparison of the criteria for updating kriging response surface models in multi-objective optimization. In: IEEE Congress on Evolutionary Computation, pp. 1–8. IEEE (2012)
Couckuyt, I., Deschrijver, D., Dhaene, T.: Fast calculation of multiobjective probability of improvement and expected improvement criteria for pareto optimization. J. Global Optim. 60(3), 575–594 (2014)
Jones, D.R., Schonlau, M., Welch, W.J.: Efficient global optimization of expensive black-box functions. J. Global Optim. 13(4), 455–492 (1998)
Mockus, J., Tiešis, V., Žilinskas, A.: The application of Bayesian methods for seeking the extremum. In: Towards Global Optimization, vol. 2, pp. 117–131. North-Holland, Amsterdam (1978)
Wagner, T., Emmerich, M., Deutz, A., Ponweiser, W.: On expected-improvement criteria for model-based multi-objective optimization. In: Schaefer, R., Cotta, C., Kołodziej, J., Rudolph, G. (eds.) PPSN 2010. LNCS, vol. 6238, pp. 718–727. Springer, Heidelberg (2010). doi:10.1007/978-3-642-15844-5_72
Emmerich, M.T., Deutz, A.H., Klinkenberg, J.W.: Hypervolume-based expected improvement: monotonicity properties and exact computation. In: IEEE Congress on Evolutionary Computation (CEC), pp. 2147–2154. IEEE (2011)
Emmerich, M., Yang, K., Deutz, A., Wang, H., Fonseca, C.M.: A multicriteria generalization of bayesian global optimization. In: Pardalos, P.M., Zhigljavsky, A., Žilinskas, J. (eds.) Advances in Stochastic and Deterministic Global Optimization. SOIA, vol. 107, pp. 229–242. Springer, Cham (2016). doi:10.1007/978-3-319-29975-4_12
Hupkens, I., Deutz, A., Yang, K., Emmerich, M.: Faster exact algorithms for computing expected hypervolume improvement. In: Gaspar-Cunha, A., Henggeler Antunes, C., Coello, C.C. (eds.) EMO 2015. LNCS, vol. 9019, pp. 65–79. Springer, Cham (2015). doi:10.1007/978-3-319-15892-1_5
Yang, K., Li, L., Deutz, A., Bäck, T., Emmerich, M.: Preference-based multiobjective optimization using truncated expected hypervolume improvement. In: 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery. IEEE (2016)
Vazquez, E., Bect, J.: Convergence properties of the expected improvement algorithm with fixed mean and covariance functions. J. Stat. Plan. Infer. 140(11), 3088–3095 (2010)
Knowles, J., Hughes, E.J.: Multiobjective optimization on a budget of 250 evaluations. In: Coello Coello, C.A., Hernández Aguirre, A., Zitzler, E. (eds.) EMO 2005. LNCS, vol. 3410, pp. 176–190. Springer, Heidelberg (2005). doi:10.1007/978-3-540-31880-4_13
Keane, A.J.: Statistical improvement criteria for use in multiobjective design optimization. AIAA J. 44(4), 879–891 (2006)
Shimoyama, K., Sato, K., Jeong, S., Obayashi, S.: Updating kriging surrogate models based on the hypervolume indicator in multi-objective optimization. J. Mech. Des. 135(9), 094503–094503-7 (2013)
Svenson, J., Santner, T.: Multiobjective optimization of expensive-to-evaluate deterministic computer simulator models. Comput. Stat. Data Anal. 94, 250–264 (2016)
Emmerich, M.T.M.: Single-and multi-objective evolutionary design optimization assisted by Gaussian random field metamodels. Ph.D. thesis, FB Informatik, University of Dortmund, ELDORADO, Dortmund, 10 (2005)
Shir, O.M., Emmerich, M., Bäckck, T., Vrakking, M.J.: The application of evolutionary multi-criteria optimization to dynamic molecular alignment. In: IEEE Congress on Evolutionary Computation, pp. 4108–4115. IEEE (2007)
Łaniewski-Wołłk, P, Obayashi S, Jeong S.: Development of expected improvement for multi-objective problems. In: Proceedings of 42nd Fluid Dynamics Conference/Aerospace Numerical, Simulation Symposium (CD ROM), Varna, Bulgaria (2010)
Luo, C., Shimoyama, K., Obayashi, S.: Kriging model based many-objective optimization with efficient calculation of expected hypervolume improvement. In: IEEE Congress on Evolutionary Computation (CEC), pp. 1187–1194. IEEE (2014)
Zitzler, E., Thiele, L.: Multiobjective evolutionary algorithms: a comparative case study and the strength pareto approach. IEEE Trans. Evol. Comput. 3(4), 257–271 (1999)
Zitzler, E., Thiele, L., Laumanns, M., Fonseca, C.M., Fonseca, V.G.D.: Performance assessment of multiobjective optimizers: an analysis and review. IEEE Trans. Evol. Comput. 7(2), 117–132 (2003)
Auger, A., Bader, J., Brockhoff, D., Zitzler, E.: Hypervolume-based multiobjective optimization: theoretical foundations and practical implications. Theor. Comput. Sci. 425, 75–103 (2012)
Emmerich, M.T.M., Fonseca, C.M.: Computing hypervolume contributions in low dimensions: asymptotically optimal algorithm and complexity results. In: Takahashi, R.H.C., Deb, K., Wanner, E.F., Greco, S. (eds.) EMO 2011. LNCS, vol. 6576, pp. 121–135. Springer, Heidelberg (2011). doi:10.1007/978-3-642-19893-9_9
Lacour, R., Klamroth, K., Fonseca, C.M.: A box decomposition algorithm to compute the hypervolume indicator. Comput. Oper. Res. 79, 347–360 (2016)
Acknowledgements
Kaifeng Yang acknowledges financial support from the China Scholarship Council (CSC), CSC No. 201306370037. Carlos M. Fonseca was supported by national funds through the Portuguese Foundation for Science and Technology (FCT), and by the European Regional Development Fund (FEDER) through COMPETE 2020 – Operational Program for Competitiveness and Internationalization (POCI).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
Definition 4
( \({\varPsi _{\infty }}\) function (see also [14])). Let \(\phi (s)= 1/\sqrt{2\pi }e^{-\frac{1}{2}s^2} (s\in \mathbb {R})\) denote the probability density function (PDF) of the standard normal distribution. Moreover, let \(\varPhi (s)= \frac{1}{2}\left( 1 + \text {erf}\left( \frac{s}{\sqrt{2}}\right) \right) \) denote its cumulative probability distribution function (CDF), and \(\text {erf}\) is Gaussian error function. The general normal distribution with mean \(\mu \) and standard deviation \(\sigma \) has as PDF, \(\xi _{\mu ,\sigma }(s)=\phi _{\mu , \sigma }(s) = \frac{1}{\sigma }\phi (\frac{s-\mu }{\sigma })\) and its CDF is \(\varPhi _{\mu , \sigma }(s) = \varPhi (\frac{s-\mu }{\sigma })\). Then the function \(\varPsi _{\infty }(a,b,\mu ,\sigma )\) is defined as:
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Yang, K., Emmerich, M., Deutz, A., Fonseca, C.M. (2017). Computing 3-D Expected Hypervolume Improvement and Related Integrals in Asymptotically Optimal Time. In: Trautmann, H., et al. Evolutionary Multi-Criterion Optimization. EMO 2017. Lecture Notes in Computer Science(), vol 10173. Springer, Cham. https://doi.org/10.1007/978-3-319-54157-0_46
Download citation
DOI: https://doi.org/10.1007/978-3-319-54157-0_46
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-54156-3
Online ISBN: 978-3-319-54157-0
eBook Packages: Computer ScienceComputer Science (R0)