A comparative study on surrogate models for SAEAs

Surrogate model assisted evolutionary algorithms (SAEAs) are metamodel-based strategies usually employed on the optimization of problems that demand a high computational cost to be evaluated. SAEAs employ metamodels, like Kriging and radial basis function (RBF), to speed up convergence towards good quality solutions and to reduce the number of function evaluations. However, investigations concerning the influence of metamodels in SAEAs performance have not been developed yet. In this context, this paper performs an investigative study on commonly adopted metamodels to compare the ordinary Kriging (OK), first-order universal Kriging (UK1), second-order universal Kriging (UK2), blind Kriging (BK) and RBF metamodels performance when embedded into a single-objective SAEA Framework (SAEA/F). The results obtained suggest that the OK metamodel presents a slightly better improvement than the others, although it does not present statistically significant difference in relation to UK1, UK2, and BK. The RBF showed the lowest computational cost, but the worst performance. However, this worse performance is around 2% in relation to the other metamodels. Furthermore, the results show that BK presents the highest computational cost without any significant improvement in solution quality when compared to OK, UK1, and UK2.

[1]  T. W. Layne,et al.  A Comparison of Approximation Modeling Techniques: Polynomial Versus Interpolating Models , 1998 .

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

[3]  Minrui Fei,et al.  Update-based evolution control: A new fitness approximation method for evolutionary algorithms , 2015 .

[4]  Rafael Martí,et al.  Experimental Testing of Advanced Scatter Search Designs for Global Optimization of Multimodal Functions , 2005, J. Glob. Optim..

[5]  Virginia Torczon,et al.  Using approximations to accelerate engineering design optimization , 1998 .

[6]  Yves Deville,et al.  DiceKriging, DiceOptim: Two R Packages for the Analysis of Computer Experiments by Kriging-Based Metamodeling and Optimization , 2012 .

[7]  Andy J. Keane,et al.  Recent advances in surrogate-based optimization , 2009 .

[8]  D. Ginsbourger,et al.  Kriging is well-suited to parallelize optimization , 2010 .

[9]  Virginia Torczon,et al.  Numerical Optimization Using Computer Experiments , 1997 .

[10]  Pramudita Satria Palar,et al.  On efficient global optimization via universal Kriging surrogate models , 2017, Structural and Multidisciplinary Optimization.

[11]  Wang Hao,et al.  Modified Sequential Kriging Optimization for Multidisciplinary Complex Product Simulation , 2010 .

[12]  Andy J. Keane,et al.  Engineering Design via Surrogate Modelling - A Practical Guide , 2008 .

[13]  Christine A. Shoemaker,et al.  Local function approximation in evolutionary algorithms for the optimization of costly functions , 2004, IEEE Transactions on Evolutionary Computation.

[14]  Ziyan Ren,et al.  Comparative Study on Kriging Surrogate Models for Metaheuristic Optimization of Multidimensional Electromagnetic Problems , 2015, IEEE Transactions on Magnetics.

[15]  Bernhard Sendhoff,et al.  Generalizing Surrogate-Assisted Evolutionary Computation , 2010, IEEE Transactions on Evolutionary Computation.

[16]  Humberto Rocha,et al.  On the selection of the most adequate radial basis function , 2009 .

[17]  Jianchao Zeng,et al.  Surrogate-Assisted Cooperative Swarm Optimization of High-Dimensional Expensive Problems , 2017, IEEE Transactions on Evolutionary Computation.

[18]  R. Storn,et al.  Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series) , 2005 .

[19]  William J. Welch,et al.  Computer experiments and global optimization , 1997 .

[20]  Agus Sudjianto,et al.  Blind Kriging: A New Method for Developing Metamodels , 2008 .

[21]  C. Koh,et al.  A Numerically Efficient Multi-Objective Optimization Algorithm: Combination of Dynamic Taylor Kriging and Differential Evolution , 2015, IEEE Transactions on Magnetics.

[22]  Bernhard Sendhoff,et al.  On Evolutionary Optimization with Approximate Fitness Functions , 2000, GECCO.

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

[24]  P. Siarry,et al.  Multiobjective Optimization: Principles and Case Studies , 2004 .

[25]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems (Genetic and Evolutionary Computation) , 2006 .

[26]  Filip De Turck,et al.  Blind Kriging: Implementation and performance analysis , 2012, Adv. Eng. Softw..

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

[28]  Kalyanmoy Deb,et al.  Multi-objective optimization using evolutionary algorithms , 2001, Wiley-Interscience series in systems and optimization.

[29]  Qingfu Zhang,et al.  A Gaussian Process Surrogate Model Assisted Evolutionary Algorithm for Medium Scale Expensive Optimization Problems , 2014, IEEE Transactions on Evolutionary Computation.

[30]  Dimitri N. Mavris,et al.  New Approaches to Conceptual and Preliminary Aircraft Design: A Comparative Assessment of a Neural Network Formulation and a Response Surface Methodology , 1998 .

[31]  Petros Koumoutsakos,et al.  Accelerating evolutionary algorithms with Gaussian process fitness function models , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[32]  Yaochu Jin,et al.  A comprehensive survey of fitness approximation in evolutionary computation , 2005, Soft Comput..

[33]  Ruichen Jin,et al.  On Sequential Sampling for Global Metamodeling in Engineering Design , 2002, DAC 2002.

[34]  Kyung K. Choi,et al.  Metamodeling Method Using Dynamic Kriging for Design Optimization , 2011 .

[35]  Andy J. Keane,et al.  Combining Global and Local Surrogate Models to Accelerate Evolutionary Optimization , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[36]  T. Simpson,et al.  Comparative studies of metamodelling techniques under multiple modelling criteria , 2001 .

[37]  Aru Yan,et al.  Synthesis of Ferromagnetic Nd 2 Fe 14 B Nanocrystalline via Solvothermal Decomposition and Reduction–Diffusion Calcination , 2015 .

[38]  T. Simpson,et al.  Use of Kriging Models to Approximate Deterministic Computer Models , 2005 .