Accelerating Evolutionary Algorithms With Gaussian Process Fitness Function Models
- 25 April 2005
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews)
- Vol. 35 (2), 183-194
- https://doi.org/10.1109/tsmcc.2004.841917
Abstract
We present an overview of evolutionary algorithms that use empirical models of the fitness function to accelerate convergence, distinguishing between evolution control and the surrogate approach. We describe the Gaussian process model and propose using it as an inexpensive fitness function surrogate. Implementation issues such as efficient and numerically stable computation, exploration versus exploitation, local modeling, multiple objectives and constraints, and failed evaluations are addressed. Our resulting Gaussian process optimization procedure clearly outperforms other evolutionary strategies on standard test functions as well as on a real-world problem: the optimization of stationary gas turbine compressor profiles.Keywords
This publication has 13 references indexed in Scilit:
- Reducing the Time Complexity of the Derandomized Evolution Strategy with Covariance Matrix Adaptation (CMA-ES)Evolutionary Computation, 2003
- Automated Design Optimization of Compressor Blades for Stationary, Large-Scale TurbomachineryPublished by ASME International ,2003
- A framework for evolutionary optimization with approximate fitness functionsIEEE Transactions on Evolutionary Computation, 2002
- Use of Adaptive Metamodeling for Design OptimizationPublished by American Institute of Aeronautics and Astronautics (AIAA) ,2002
- Adapting Weighted Aggregation for Multiobjective Evolution StrategiesLecture Notes in Computer Science, 2001
- A rigorous framework for optimization of expensive functions by surrogatesStructural and Multidisciplinary Optimization, 1999
- Using approximations to accelerate engineering design optimizationPublished by American Institute of Aeronautics and Astronautics (AIAA) ,1998
- Comparison of response surface and kriging models for multidisciplinary design optimizationPublished by American Institute of Aeronautics and Astronautics (AIAA) ,1998
- Accelerating the convergence of evolutionary algorithms by fitness landscape approximationLecture Notes in Computer Science, 1998
- Minimax and maximin distance designsJournal of Statistical Planning and Inference, 1990