Evolutionary optimization for computationally expensive problems using Gaussian processes

The use of statistical models to approximate detailed analysis codes for evolutionary optimization has attracted some attention [1-3]. However, those early methodologies do suffer from some limitations, the most serious of which being the extra tuning parameter introduceds. Also the question of when to include more data points to the approximation model during the search remains unresolved. Those limitations might seriously impede their successful application. We present here an approach that makes use of the extra information provided by a Gaussian processes (GP) approximation model to guide the crucial model update step. We present here the advantages of using GP over other neural-net biologically inspired approaches. Results are presented for a real world-engineering problem involving the structural optimization of a satellite boom.

[1]  Andy J. Keane,et al.  Metamodeling Techniques For Evolutionary Optimization of Computationally Expensive Problems: Promises and Limitations , 1999, GECCO.

[2]  Andy J. Keane,et al.  Energy flow predictions in a structure of rigidly joined beams using receptance theory , 1995 .

[3]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1992, Artificial Intelligence.

[4]  W. Press,et al.  Numerical Recipes: The Art of Scientific Computing , 1987 .

[5]  P. A. Newman,et al.  Optimization with variable-fidelity models applied to wing design , 1999 .

[6]  Timothy W. Simpson,et al.  On the Use of Statistics in Design and the Implications for Deterministic Computer Experiments , 1997 .

[7]  Vassili Toropov,et al.  Multiparameter structural optimization using FEM and multipoint explicit approximations , 1993 .

[8]  J. Sobieszczanski-Sobieski.  Optimization by decomposition , 1993 .

[9]  Jaroslaw Sobieszczanski-Sobieski,et al.  Multidisciplinary aerospace design optimization - Survey of recent developments , 1996 .

[10]  Vassili Toropov,et al.  Simulation approach to structural optimization , 1989 .

[11]  Prabhat Hajela,et al.  Topological optimization of rotorcraft subfloor structures for crashworthiness considerations , 1997 .

[12]  Andy J. Keane,et al.  Passive vibration control via unusual geometries: experiments on model aerospace structures , 1996 .

[13]  J. -F. M. Barthelemy,et al.  Approximation concepts for optimum structural design — a review , 1993 .

[14]  Zbigniew Michalewicz,et al.  Genetic algorithms + data structures = evolution programs (3rd ed.) , 1996 .

[15]  Geoffrey E. Hinton,et al.  Bayesian Learning for Neural Networks , 1995 .

[16]  Alain Ratle,et al.  Accelerating the Convergence of Evolutionary Algorithms by Fitness Landscape Approximation , 1998, PPSN.

[17]  A. Ratle.  Optimal sampling strategies for learning a fitness model , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).