Abstract
In this article, we present a framework for taking into account user preferences in multi-objective Bayesian optimization in the case where the objectives are expensive-to-evaluate black-box functions. A novel expected improvement criterion to be used within Bayesian optimization algorithms is introduced. This criterion, which we call the expected weighted hypervolume improvement (EWHI) criterion, is a generalization of the popular expected hypervolume improvement to the case where the hypervolume of the dominated region is defined using a user-defined absolutely continuous measure instead of the Lebesgue measure. The EWHI criterion takes the form of an integral for which no closed form expression exists in the general case. To deal with its computation, we propose an importance sampling approximation method. A sampling density that is optimal for the computation of the EWHI for a predefined set of points is crafted and a sequential Monte-Carlo (SMC) approach is used to obtain a sample approximately distributed from this density. The ability of the criterion to produce optimization strategies oriented by user preferences is demonstrated on a simple bi-objective test problem in the cases of a preference for one objective and of a preference for certain regions of the Pareto front.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In the original definition, the authors introduce additional terms to weight the axis. In this work, one of our objective is to get rid of the bounding set , as proposed by [13]. Therefore we do not consider these terms.
References
Astudillo, R., Frazier, P.: Multi-attribute Bayesian optimization under utility uncertainty. In: Proceedings of the NIPS Workshop on Bayesian Optimization, December 2017, Long Beach, USA (To appear)
Au, S.K., Beck, J.L.: Estimation of small failure probabilities in high dimensions by subset simulation. Probab. Eng. Mech. 16(4), 263–277 (2001)
Auger, A., Bader, J., Brockhoff, D., Zitzler, E.: Articulating user preferences in many-objective problems by sampling the weighted hypervolume. In: Proceedings of the 11th Annual Conference on Genetic and Evolutionary Computation, pp. 555–562. ACM (2009)
Auger, A., Bader, J., Brockhoff, D., Zitzler, E.: Investigating and exploiting the bias of the weighted hypervolume to articulate user preferences. In: Proceedings of the 11th Annual Conference on Genetic and Evolutionary Computation, pp. 563–570. ACM (2009)
Bect, J., Ginsbourger, D., Li, L., Picheny, V., Vazquez, E.: Sequential design of computer experiments for the estimation of a probability of failure. Stat. Comput. 22(3), 773–793 (2012)
Bect, J., Li, L., Vazquez, E.: Bayesian subset simulation. SIAM/ASA J. Uncertain. Quantif. 5(1), 762–786 (2017)
Bect, J., Sueur, R., Gérossier, A., Mongellaz, L., Petit, S., Vazquez, E.: Échantillonnage préférentiel et méta-modèles: méthodes bayésiennes optimale et défensive. In: 47èmes Journées de Statistique de la SFdS-JdS 2015 (2015)
Benassi, R.: Nouvel algorithme d’optimisation bayésien utilisant une approche Monte-Carlo séquentielle. Ph.D. thesis, Supélec (2013)
Cérou, F., Del Moral, P., Furon, T., Guyader, A.: Sequential Monte Carlo for rare event estimation. Stat. Comput. 22(3), 795–808 (2012)
Chafekar, D., Xuan, J., Rasheed, K.: Constrained multi-objective optimization using steady state genetic algorithms. In: Cantú-Paz, E., et al. (eds.) GECCO 2003. LNCS, vol. 2723, pp. 813–824. Springer, Heidelberg (2003). https://doi.org/10.1007/3-540-45105-6_95
Del Moral, P., Doucet, A., Jasra, A.: Sequential monte carlo samplers. J. R. Stat. Soc.: Ser. B (Stat. Methodol.) 68(3), 411–436 (2006)
Emmerich, M.: Single - and multiobjective evolutionary design optimization assisted by Gaussian random field metamodels. Ph.D. thesis, Technical University Dortmund (2005)
Emmerich, M., Deutz, A.H., Yevseyeva, I.: On reference point free weighted hypervolume indicators based on desirability functions and their probabilistic interpretation. Proc. Technol. 16, 532–541 (2014)
Emmerich, M., Giannakoglou, K.C., Naujoks, B.: Single - and multi-objective evolutionary optimization assisted by Gaussian random field metamodels. IEEE Trans. Evol. Comput. 10(4), 421–439 (2006)
Emmerich, M., Klinkenberg, J.W.: The computation of the expected improvement in dominated hypervolume of Pareto front approximations. Technical report, Leiden University (2008)
Feliot, P.: A Bayesian approach to constrained multi-objective optimization. Ph.D. thesis, IRT SystemX and Centrale Supélec (2017)
Feliot, P., Bect, J., Vazquez, E.: A Bayesian approach to constrained single-and multi-objective optimization. J. Glob. Optim. 67(1–2), 97–133 (2017)
Ginsbourger, D., Le Riche, R.: Towards Gaussian process-based optimization with finite time horizon. In: Giovagnoli, A., Atkinson, A., Torsney, B., May, C. (eds.) mODa 9 – Advances in Model-Oriented Design and Analysis. Contributions to Statistics, pp. 89–96. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-7908-2410-0_12
Harrington, E.C.: The desirability function. Ind. Qual. Control. 21(10), 494–498 (1965)
Jones, D.R., Schonlau, M., Welch, W.J.: Efficient global optimization of expensive black-box functions. J. Glob. Optim. 13(4), 455–492 (1998)
Knowles, J., Corne, D.: On metrics for comparing nondominated sets. In: Proceedings of the 2002 Congress on Evolutionary Computation, CEC 2002, vol. 1, pp. 711–716. IEEE (2002)
Kushner, H.J.: A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise. J. Fluids Eng. 86(1), 97–106 (1964)
Laumanns, M., Rudolph, G., Schwefel, H.P.: Approximating the pareto set: concepts, diversity issues, and performance assessment. Secretary of the SFB 531 (1999)
Mockus, J., Tiesis, V., Žilinskas, A.: The application of Bayesian methods for seeking the extremum. In: Dixon, L.C.W., Szegö, G.P. (eds.) Towards Global Optimization, vol. 2, pp. 117–129. North Holland, New York (1978)
Santner, T.J., Williams, B.J., Notz, W.: The Design and Analysis of Computer Experiments. Springer Series in Statistics. Springer, New York (2003). https://doi.org/10.1007/978-1-4757-3799-8
Schonlau, M., Welch, W.J., Jones, D.R.: Global versus local search in constrained optimization of computer models. In: New Developments and Applications in Experimental Design: Selected Proceedings of a 1997 Joint AMS-IMS-SIAM Summer Conference. IMS Lecture Notes-Monographs Series, vol. 34, pp. 11–25. Institute of Mathematical Statistics (1998)
Wagner, T., Trautmann, H.: Integration of preferences in hypervolume-based multiobjective evolutionary algorithms by means of desirability functions. IEEE Trans. Evol. Comput. 14(5), 688–701 (2010)
Williams, C.K.I., Rasmussen, C.: Gaussian Processes for Machine Learning, vol. 2(3), p. 4. The MIT Press, Cambridge (2006)
Zitzler, E., Brockhoff, D., Thiele, L.: The hypervolume indicator revisited: on the design of pareto-compliant indicators via weighted integration. In: Obayashi, S., Deb, K., Poloni, C., Hiroyasu, T., Murata, T. (eds.) EMO 2007. LNCS, vol. 4403, pp. 862–876. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-70928-2_64
Zitzler, E., Thiele, L.: Multiobjective optimization using evolutionary algorithms — a comparative case study. In: Eiben, A.E., Bäck, T., Schoenauer, M., Schwefel, H.-P. (eds.) PPSN 1998. LNCS, vol. 1498, pp. 292–301. Springer, Heidelberg (1998). https://doi.org/10.1007/BFb0056872
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Approximate Variance of the EI Estimator
A Approximate Variance of the EI Estimator
We derive in this appendix the variance of the SMC estimator for \(\rho _n\). In the SMC procedure that we consider, the particles \(\left( y_{n,i}\right) _{1 \le i \le m}\) are obtained from a sequence of densities \((\pi _{n,t})_{0 \le t \le T}\), where \(\pi _{n,0}\) is an easy-to-sample initial density and \(\pi _{n,T} = \pi _n\) is the target density. Let \((\gamma _{n,t})_{0 \le t \le T}\) and \((Z_{n,t})_{0 \le t \le T}\) denote the corresponding sequences of un-normalized densities and normalizing constants.
First, observe that, for \(1\le t \le T\),
Thus, we can derive a sequence of approximations \(\widehat{Z}_{n,t}\) of \(Z_{n,t}\), \(t\ge 1\), using the following recursion formula:
where the particles \((y_{n,t-1,i})_{1 \le i \le m}\sim \pi _{n,t-1}\) are obtained using an SMC procedure (see, e.g., [6]). The estimator of \(\rho _n(x)\) that we actually consider is then
where
and
with
Now, assume the idealized setting, as usual in the SMC literature (see, e.g., [9]), where
-
(i)
\(y_{n,t,i} {\mathop {\sim }\limits ^{\scriptstyle \text {i.i.d}}} \pi _{n,t}\), \(1 \le i \le m\),
-
(ii)
the samples \(\mathcal {Y}_{n,t} = (y_{n,t,i})_{1 \le i \le m}\) are independent, \(0\le t \le T\).
Observe from (19) and (24) that under (i), \(\widehat{\alpha }_n(x)\) is an unbiased estimator of \(\alpha _n(x) = \frac{\rho _n(x)}{Z_n}\), and \(\widehat{\theta }_{n,t}\) is an unbiased estimator of \(\theta _{n,t} = \frac{Z_{n,t}}{Z_{n,t-1}}\), \(1 \le t \le T\). Moreover, under (ii), \(\widehat{\alpha }_n(x)\) and the \((\widehat{\theta }_{n,t})_{1 \le t \le T}\) are independent. Thus,
We obtain the coefficient of variation of \(\widehat{\rho }_n(x)\)
where \(\varLambda _n(x)^{2} = \frac{\mathrm {Var}\,\widehat{\alpha }_n(x)}{\alpha _n(x)^2}\) and \(\varDelta _{n,t}^2 = \frac{\mathrm {Var} \widehat{Z}_{n,t}}{Z_{n,t}^2}\) are the coefficients of variation of \(\widehat{\alpha }_n(x)\) and \(\widehat{Z}_{n,t}\) respectively.
Using the same ideas as above, we have
where \(\delta _{n,t}^2 = \frac{\mathrm {Var}\, \widehat{\theta }_{n,t}}{\theta _{n,t}^2}\) is the coefficient of variation of \(\widehat{\theta }_{n,t}\).
Estimators of \(\varLambda _{n}(x)^2\), \(\varDelta _{n,t}^2\) and \(\delta _{n,t}^2\) can be derived under (ii). For instance, observe that
Thus, an estimator of \(\delta _{n,t}^2\) is
Plugging (33) in (31), we obtain an estimator of \(\varDelta _{n,t}^2\):
Similarly, an estimator of \(\varLambda _n(x)^2\) is
As a result, we obtain the following numerically tractable approximation of the variance of \(\widehat{\rho }_n(x)\):
where \(\widehat{Z}_{n,t}\) and \(\widehat{\varDelta }_{n,t}^2\) are obtained recursively using (25) and (34), \(\widehat{\varLambda }_n(x)^2\) is computed using (35) and \(\widehat{\rho }_n(x)\) is computed using (26).
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Feliot, P., Bect, J., Vazquez, E. (2019). User Preferences in Bayesian Multi-objective Optimization: The Expected Weighted Hypervolume Improvement Criterion. In: Nicosia, G., Pardalos, P., Giuffrida, G., Umeton, R., Sciacca, V. (eds) Machine Learning, Optimization, and Data Science. LOD 2018. Lecture Notes in Computer Science(), vol 11331. Springer, Cham. https://doi.org/10.1007/978-3-030-13709-0_45
Download citation
DOI: https://doi.org/10.1007/978-3-030-13709-0_45
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-13708-3
Online ISBN: 978-3-030-13709-0
eBook Packages: Computer ScienceComputer Science (R0)