Accelerating evolutionary algorithms with Gaussian process fitness function models
Date
2005
Authors
Buche, Dirk
Schraudolph, Nicol
Koumoutsakos, Petros
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers (IEEE Inc)
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.
Description
Keywords
Keywords: Compressors; Computation theory; Control system synthesis; Convergence of numerical methods; Functions; Gas turbines; Mathematical models; Optimization; Evolution control; Fitness function modeling; Gas turbine compressor design; Gaussian process; Surroga Evolution control; Evolutionary algorithms (EAs); Fitness function modeling; Gas turbine compressor design; Gaussian process; Surrogate approach
Citation
Collections
Source
IEEE Transactions on Systems, Man and Cybernetics. Part C: Applications and Reviews
Type
Journal article