A framework for evolutionary optimization with approximate fitness functions

It is not unusual that an approximate model is needed for fitness evaluation in evolutionary computation. In this case, the convergence properties of the evolutionary algorithm are unclear due to the approximation error of the model. In this paper, extensive empirical studies are carried out to investigate the convergence properties of an evolution strategy using an approximate fitness function on two benchmark problems. It is found that incorrect convergence will occur if the approximate model has false optima. To address this problem, individual- and generation-based evolution control are introduced and the resulting effects on the convergence properties are presented. A framework for managing approximate models in generation-based evolution control is proposed. This framework is well suited for parallel evolutionary optimization, which is able to guarantee the correct convergence of the evolutionary algorithm, as well as to reduce the computation cost as much as possible. Control of the evolution and updating of the approximate models are based on the estimated fidelity of the approximate model. Numerical results are presented for three test problems and for an aerodynamic design example.

[1]  H. H. Rosenbrock,et al.  An Automatic Method for Finding the Greatest or Least Value of a Function , 1960, Comput. J..

[2]  Jorge J. Moré,et al.  Recent Developments in Algorithms and Software for Trust Region Methods , 1982, ISMP.

[3]  D. Ackley.  A connectionist machine for genetic hillclimbing , 1987 .

[4]  Karl Sims,et al.  Artificial evolution for computer graphics , 1991, SIGGRAPH.

[5]  Stuart A. Kauffman,et al.  The origins of order , 1993 .

[6]  Kalyanmoy Deb,et al.  Genetic Algorithms, Noise, and the Sizing of Populations , 1992, Complex Syst..

[7]  David J. C. MacKay,et al.  Information-Based Objective Functions for Active Data Selection , 1992, Neural Computation.

[8]  Toshiyuki Masui,et al.  Graphic object layout with interactive genetic algorithms , 1992, Proceedings IEEE Workshop on Visual Languages.

[9]  Jochem Zowe,et al.  A Version of the Bundle Idea for Minimizing a Nonsmooth Function: Conceptual Idea, Convergence Analysis, Numerical Results , 1992, SIAM J. Optim..

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

[11]  W. Carpenter,et al.  A comparison of polynomial approximations and artificial neural nets as response surfaces , 1993 .

[12]  D. Grierson,et al.  Optimal sizing, geometrical and topological design using a genetic algorithm , 1993 .

[13]  Mark Plutowski,et al.  Selecting concise training sets from clean data , 1993, IEEE Trans. Neural Networks.

[14]  John A. Biles,et al.  GenJam: A Genetic Algorithm for Generating Jazz Solos , 1994, ICMC.

[15]  Nikolaus Hansen,et al.  A Derandomized Approach to Self-Adaptation of Evolution Strategies , 1994, Evolutionary Computation.

[16]  Jongsoo Lee,et al.  Genetic algorithms in multidisciplinary rotor blade design , 1995 .

[17]  Erick Cantú-Paz,et al.  A summary of Research on Parallel Genetic Algorithms , 1995 .

[18]  David B. Fogel,et al.  Evolutionary Computation: Towards a New Philosophy of Machine Intelligence , 1995 .

[19]  R. H. Myers,et al.  Response Surface Methodology: Process and Product Optimization Using Designed Experiments , 1995 .

[20]  Thomas Bäck,et al.  Evolutionary Algorithms in Theory and Practice , 1996 .

[21]  William E. Hart,et al.  Optimization with genetic algorithm hybrids that use local searches , 1996 .

[22]  Nikolaus Hansen,et al.  Adapting arbitrary normal mutation distributions in evolution strategies: the covariance matrix adaptation , 1996, Proceedings of IEEE International Conference on Evolutionary Computation.

[23]  J. Redmond,et al.  Actuator placement based on reachable set optimization for expected disturbance , 1996 .

[24]  Natalia Alexandrov,et al.  Multidisciplinary design optimization : state of the art , 1997 .

[25]  David B. Fogel,et al.  Evolutionary algorithms in theory and practice , 1997, Complex.

[26]  Les A. Piegl,et al.  The NURBS Book , 1995, Monographs in Visual Communication.

[27]  Thomas Bck,et al.  Evolutionary computation: Toward a new philosophy of machine intelligence , 1997, Complex..

[28]  Timothy M. Mauery,et al.  COMPARISON OF RESPONSE SURFACE AND KRIGING MODELS FOR MULTIDISCIPLINARY DESIGN OPTIMIZATION , 1998 .

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

[30]  A. J. Booker,et al.  A rigorous framework for optimization of expensive functions by surrogates , 1998 .

[31]  Stephane Pierret,et al.  Turbomachinery Blade Design Using a Navier-Stokes Solver and Artificial Neural Network , 1998 .

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

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

[34]  Wei Shyy,et al.  Response surface and neural network techniques for rocket engine injector optimization , 1999 .

[35]  Carlos A. Coello Coello,et al.  An updated survey of evolutionary multiobjective optimization techniques: state of the art and future trends , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[36]  Sethu Vijayakumar,et al.  Improving Generalization Ability through Active Learning , 1999 .

[37]  Edmund K. Burke,et al.  A multi-stage approach for the thermal generator maintenance scheduling problem , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[38]  Larry Bull,et al.  On Model-Based Evolutionary Computation , 1999, Soft Comput..

[39]  D. Keymeulen,et al.  Evolutionary design of electronic devices and circuits , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

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

[41]  X. Yao.  Evolutionary Search of Approximated N-dimensional Landscapes , 2000 .

[42]  David W. Corne,et al.  Heuristics for Evolutionary Off-line Routing in Telecommunication Networks , 2000, GECCO.

[43]  Bernhard Sendhoff,et al.  Optimisation of a Stator Blade Used in a Transonic Compressor Cascade with Evolution Strategies , 2000 .

[44]  Kenji Fukumizu,et al.  Statistical active learning in multilayer perceptrons , 2000, IEEE Trans. Neural Networks Learn. Syst..

[45]  Christian Igel,et al.  Evolutionary Parameter Optimization for Visual Obstacle Detection , 2000 .

[46]  Yaochu Jin,et al.  Managing approximate models in evolutionary aerodynamic design optimization , 2001, Proceedings of the 2001 Congress on Evolutionary Computation (IEEE Cat. No.01TH8546).

[47]  Nikolaus Hansen,et al.  Completely Derandomized Self-Adaptation in Evolution Strategies , 2001, Evolutionary Computation.

[48]  Thomas J. Santner,et al.  Design and analysis of computer experiments , 1998 .

[49]  J. H. North,et al.  Design by natural selection , 2004, Synthese.