User Preferences in Bayesian Multi-objective Optimization: The Expected Weighted Hypervolume Improvement Criterion

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.

[1]  Thomas Bäck,et al.  Parallel Problem Solving from Nature — PPSN V , 1998, Lecture Notes in Computer Science.

[2]  Anne Auger,et al.  Investigating and exploiting the bias of the weighted hypervolume to articulate user preferences , 2009, GECCO.

[3]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[4]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[5]  Michael T. M. Emmerich,et al.  Single- and multiobjective evolutionary optimization assisted by Gaussian random field metamodels , 2006, IEEE Transactions on Evolutionary Computation.

[6]  Thomas J. Santner,et al.  The Design and Analysis of Computer Experiments , 2003, Springer Series in Statistics.

[7]  Lothar Thiele,et al.  Multiobjective Optimization Using Evolutionary Algorithms - A Comparative Case Study , 1998, PPSN.

[8]  Heike Trautmann,et al.  Integration of Preferences in Hypervolume-Based Multiobjective Evolutionary Algorithms by Means of Desirability Functions , 2010, IEEE Transactions on Evolutionary Computation.

[9]  Iryna Yevseyeva,et al.  On Reference Point Free Weighted Hypervolume Indicators based on Desirability Functions and their Probabilistic Interpretation , 2014 .

[10]  Ling Li,et al.  Bayesian Subset Simulation , 2016, SIAM/ASA J. Uncertain. Quantification.

[11]  Julien Bect,et al.  Échantillonnage préférentiel et méta-modèles : méthodes bayésiennes optimale et défensive , 2015 .

[12]  Romain Benassi,et al.  Nouvel algorithme d'optimisation bayésien utilisant une approche Monte-Carlo séquentielle. , 2013 .

[13]  Khaled Rasheed,et al.  Constrained Multi-objective Optimization Using Steady State Genetic Algorithms , 2003, GECCO.

[14]  Harold J. Kushner,et al.  A New Method of Locating the Maximum Point of an Arbitrary Multipeak Curve in the Presence of Noise , 1964 .

[15]  Donald R. Jones,et al.  Global versus local search in constrained optimization of computer models , 1998 .

[16]  J. Beck,et al.  Estimation of Small Failure Probabilities in High Dimensions by Subset Simulation , 2001 .

[17]  William T. Scherer,et al.  "The desirability function: underlying assumptions and application implications" , 1998, SMC'98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.98CH36218).

[18]  D. Ginsbourger,et al.  Towards Gaussian Process-based Optimization with Finite Time Horizon , 2010 .

[19]  Ling Li,et al.  Sequential design of computer experiments for the estimation of a probability of failure , 2010, Statistics and Computing.

[20]  Anne Auger,et al.  Articulating user preferences in many-objective problems by sampling the weighted hypervolume , 2009, GECCO.

[21]  P. Moral,et al.  Sequential Monte Carlo samplers , 2002, cond-mat/0212648.

[22]  Julien Bect,et al.  A Bayesian Approach to Constrained Multi-objective Optimization , 2015, LION.

[23]  Pierre Del Moral,et al.  Sequential Monte Carlo for rare event estimation , 2012, Stat. Comput..

[24]  H. Schwefel,et al.  Approximating the Pareto Set: Concepts, Diversity Issues, and Performance Assessment , 1999 .

[25]  Raul Astudillo.  Multi-Attribute Bayesian Optimization under Utility Uncertainty , 2017 .

[26]  M. Emmerich,et al.  The computation of the expected improvement in dominated hypervolume of Pareto front approximations , 2008 .

[27]  Lothar Thiele,et al.  The Hypervolume Indicator Revisited: On the Design of Pareto-compliant Indicators Via Weighted Integration , 2007, EMO.

[28]  Joshua D. Knowles,et al.  On metrics for comparing nondominated sets , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).