Abstract
Prior to introducing the particular algorithms in Sect. 2.2, the more general foundations of evolution strategies are introduced in Sect. 2.1. To start with, the definition of an optimization task as used throughout this book is given in Sect. 2.1.1. Following [58], Sect. 2.1.2 presents a discussion of evolution strategy metaheuristics as a special case of evolutionary algorithms.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
This statement, however, is not meant to support the myth mentioned explicitly by Rudolph [58]: “Since early theoretical publications mainly analyzed simple ES without recombination, somehow the myth arose that ES put more emphasis on mutation than on recombination: This is a fatal misconception! Recombination has been an important ingredient of ES from the early beginning and this is still valid today.”
- 2.
See Sect. 12.2.1 in [17] for the definition of a distance measure.
- 3.
In the case of the (1+1)-ES the strategy parameters may be assigned to the algorithm itself instead of the individual, because only one set of strategy parameters is needed. This also holds for any strategy parameters which are not needed on the individual level (for example the covariance matrix of the CMA-ES).
- 4.
Algorithm 3 in [58].
- 5.
The normal distribution achieves maximum entropy among the distributions on the real domain. (See [64] for more details.)
- 6.
A symmetric matrix \(\mathbf{A} \in {\mathbb{R}}^{n\times n}\) is positive definite iff \({\mathbf{x}}^{T}\mathbf{A}\mathbf{x} > 0\) for all \(\mathbf{x} \in {\mathbb{R}}^{n}\setminus \{\mathbf{0}\}\) [17].
- 7.
For an orthogonal matrix A, \(\mathbf{A}{\mathbf{A}}^{T} ={ \mathbf{A}}^{T}\mathbf{A} = \mathbf{I}\) holds.
- 8.
See Sect. 6.2.2.3 in [17].
- 9.
The rectangular corridor model according to [8]: \(f_{1}(\mathbf{x}) = c_{0} + c_{1} \cdot x_{1}\) if the constraints \(g_{j}(\mathbf{x}): x_{j} \leq b\) with \(b \in {\mathbb{R}}^{+}\) for \(j \in \{2,\ldots,n\}\) are fulfilled, f 1(x) = ∞ otherwise.
- 10.
The sphere model according to [8]: \(f_{2}(\mathbf{x}) = c_{0} + c_{1} \cdot \sum _{n}^{i=1}{(x_{i} - x_{i}^{{\ast}})}^{2}\).
- 11.
The exact values are 0.184 and 0.2025 for the corridor and sphere models, respectively [8].
- 12.
MSC is an abbreviation of mutative self-adaptation of covariances.
- 13.
In the original publication it is called (1,λ)-ES with derandomized mutative step size.
- 14.
This way, adapting the step size by a factor ξ requires at least 1∕β > 1 generations.
- 15.
In the original paper, the algorithm is called (1,λ)-ES with derandomized mutative step size control using accumulated information.
- 16.
The column vectors of the matrix B form a so-called generating set, which motivates the terminology generating set adaptation.
- 17.
According to [32], the suggestion to use weighted recombination within the CMA-ES is due to Ingo Rechenberg, based on personal communication in 1998.
- 18.
See [17]: \(\Gamma (n) =\int _{ 0}^{\infty }{x}^{n-1}\exp (-x)\,\mbox{ d}x\).
- 19.
With the additional condition for A to consist of at least m = n 2 tuples.
- 20.
Compare Sect. 19.2.1.2 in [17].
- 21.
The term active is motivated by the fact that specifically the bad offspring individuals play an active role, although they would normally not be taken into account after selection has been applied.
- 22.
This is explicitly avoided due to the occurrence of numerical instabilities for certain objective functions; see [40].
- 23.
Population sizes μ and λ are not counted.
- 24.
Instead of the term symmetrical, this is called mirrored in the context of this strategy.
- 25.
In [26] the eNES are called exact natural evolution strategies.
- 26.
- 27.
See [70] for literature references on these topics as well as the Kriging modeling method.
- 28.
For λ def the standard setting of a (μ W ,λ)-CMA-ES with \(\lambda _{\mathit{def }} = 4 + \lfloor 3\log n\rfloor \) is used.
- 29.
In principal, any modeling technique can be used to establish the relationship between the exogenous parameters and the performance measure.
- 30.
For example the NP-hard Traveling Salesman Problem.
Bibliography
S. Amari, Natural gradient works efficiently in learning. Neural Comput. 10(2), 251–276 (1998)
D.V. Arnold, N. Hansen, Active covariance matrix adaptation for the (1+1)-CMA-ES, in Proceedings of the 12th Annual Conference on Genetic and Evolutionary Computation (GECCO’10), Portland, ed. by M. Pelikan, J. Branke (ACM, New York, 2010), pp. 385–392
D.V. Arnold, R. Salomon, Evolutionary gradient search revisited. IEEE Trans. Evol. Comput. 11(4), 480–495 (2007)
A. Auger, N. Hansen, Performance evaluation of an advanced local search evolutionary algorithm, in Proceedings of the IEEE Congress on Evolutionary Computation (CEC’05), Edinburgh, vol. 2, ed. by B. McKay et al. (IEEE, Piscataway, 2005), pp. 1777–1784
A. Auger, N. Hansen, A restart CMA evolution strategy with increasing population size, in Proceedings of the IEEE Congress on Evolutionary Computation (CEC’05), Edinburgh, vol. 2, ed. by B. McKay et al. (IEEE, Piscataway, 2005), pp. 1769–1776
A. Auger, M. Schoenauer, N. Vanhaecke, LS-CMA-ES: a second-order algorithm for covariance matrix adaptation, in Proceedings of the 8th International Conference on Parallel Problem Solving from Nature (PPSN VIII), Birmingham, ed. by X. Yao et al. Volume 3242 of Lecture Notes in Computer Science (Springer, Berlin, 2004), pp. 182–191
A. Auger, D. Brockhoff, N. Hansen, Mirrored sampling in evolution strategies with weighted recombination, in Proceedings of the 13th Annual Genetic and Evolutionary Computation Conference (GECCO’11), Dublin, ed. by N. Krasnogor, P.L. Lanzi (ACM, New York, 2011), pp. 861–868
T. Bäck, Evolutionary Algorithms in Theory and Practice (Oxford University Press, New York, 1996)
T. Bartz-Beielstein, C. Lasarczyk, M. Preuss, Sequential parameter optimization, in Proceedings of the IEEE Congress on Evolutionary Computation (CEC’05), Edinburgh, ed. by B. McKay et al. (IEEE, Piscataway, 2005), pp. 773–780
N. Beume, B. Naujoks, M. Emmerich, SMS-EMOA: multiobjective selection based on dominated hypervolume. Eur. J. Oper. Res. 181, 1653–1669 (2007)
H.-G. Beyer, B. Sendhoff, Covariance matrix adaptation revisited – the CMSA evolution strategy, in Proceedings of the 10th International Conference on Parallel Problem Solving from Nature (PPSN X), Dortmund, ed. by G. Rudolph et al. Volume 5199 in Lecture Notes in Computer Science (Springer, Berlin, 2008), pp. 123–132
Z. Bouzarkouna, A. Auger, D.-Y. Ding, Local-meta-model CMA-ES for partially separable functions, in Proceedings of the 13th Annual Genetic and Evolutionary Computation Conference (GECCO’11), Dublin, ed. by N. Krasnogor et al. (ACM, New York, 2011), pp. 869–876
D. Brockhoff, A. Auger, N. Hansen, D.V. Arnold, T. Hohm, Mirrored sampling and sequential selection for evolution strategies, in Proceedings of the 11th International Conference on Parallel Problem Solving from Nature (PPSN XI), Kraków, ed. by R. Schaefer et al. Volume 6238 in Lecture Notes in Computer Science. (Springer, Berlin, 2010), pp. 11–21
I.N. Bronstein, K.A. Semendjajew, G. Musiol, H. Muehlig, Taschenbuch der Mathematik, 7th edn. (Harri Deutsch, Frankfurt am Main, 2008)
C.A. Coello Coello, Constraint-handling techniques used with evolutionary algorithms, in Proceedings of the 13th Annual Genetic and Evolutionary Computation Conference (GECCO’11), Companion Material, Dublin, ed. by N. Krasnogor et al. (ACM, New York, 2011), pp. 1137–1160
K. Deb, Multiobjective Optimization Using Evolutionary Algorithms. Wiley-Interscience Series in Systems and Optimization (Wiley, Chichester, 2001)
K. Deb, A. Pratap, S. Agarwal, T. Meyarivan, A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans. Evol. Comput. 6(2), 182–197 (2002)
A.E. Eiben, J.E. Smith, Introduction to Evolutionary Computing. Natural Computing Series (Springer, Berlin, 2003)
T. Glasmachers, T. Schaul, Y. Sun, D. Wierstra, J. Schmidhuber, Exponential natural evolution strategies, in Proceedings of the 12th Annual Conference on Genetic and Evolutionary Computation (GECCO’10), Portland, ed. by M. Pelikan, J. Branke (ACM, New York, 2010)
D.E. Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning (Addison-Wesley, Boston, 1989)
W.H. Greene, Econometric Analysis, 4th edn. (Prentice Hall, Upper Saddle River, 1997)
N. Hansen, The CMA evolution strategy: a tutorial. Continuously updated technical report, available via http://www.lri.fr/~hansen/cmatutorial.pdf. Accessed 12 Mar 2011
N. Hansen, S. Kern, Evaluating the CMA evolution strategy on multimodal test functions, in Proceedings of the 9th International Conference on Parallel Problem Solving from Nature (PPSN VIII), Birmingham. Volume 3242 of Lecture Notes in Computer Science, ed. by X. Yao et al. (Springer, 2004), pp. 282–291
N. Hansen, A. Ostermeier, Adapting arbitrary normal mutation distributions in evolution strategies: the covariance matrix adaptation, in Proceedings of the 1996 IEEE International Conference on Evolutionary Computation (ICEC’96), Nagoya, ed. by Y. Davidor et al. (IEEE, Piscataway, 1996), pp. 312–317
N. Hansen, A. Ostermeier, Completely derandomized self-adaptation in evolution strategies. Evol. Comput. 9(2), 159–195 (2001)
N. Hansen, A. Ostermeier, A. Gawelczyk, On the adaptation of arbitrary normal mutation distributions in evolution strategies: the generating set adaptation, in Proceedings of the 6th International Conference on Genetic Algorithms (ICGA 6), Pittsburgh, ed. by L.J. Eshelman (Morgan Kaufmann, San Francisco, 1995), pp. 57–64
N. Hansen, A. Auger, R. Ros, S. Finck, P. Posik, Comparing results of 31 algorithms from the black-box optimization benchmarking BBOB-2009, in Proceedings of the 12th International Conference on Genetic and Evolutionary Computation Conference (GECCO’10), Companion Material, Portland, ed. by M. Pelikan, J. Branke (ACM, New York, 2010), pp. 1689–1696
C. Igel, T. Suttorp, N. Hansen, A computational efficient covariance matrix update and a (1+1)-CMA for evolution strategies, in Proceedings of the 8th Annual Conference on Genetic and Evolutionary Computation (GECCO’06), Seattle, ed. by M. Keijzer et al. (ACM, New York, 2006), pp. 453–460
G.A. Jastrebski, Improving evolution strategies through active covariance matrix adaptation. Master’s thesis, Faculty of Computer Science, Dalhousie University, 2005
G.A. Jastrebski, D.V. Arnold, Improving evolution strategies through active covariance matrix adaptation, in Proceedings of the 2006 IEEE Congress on Evolutionary Computation (CEC’06), Vancouver, BC, Canada, ed. by G.G. Yen et al. (IEEE, Piscataway, 2006), pp. 2814–2821
O. Kramer, A review of constraint-handling techniques for evolution strategies. Appl Comput. Int. Soft Comput. 2010, 1–11 (2010)
R. Li, Mixed-integer evolution strategies for parameter optimization and their applications to medical image analysis. PhD thesis, Leiden Institute of Advanced Computer Science (LIACS), Faculty of Science, Leiden University, 2009
D.G. Luenberger, Y. Ye, Linear and Nonlinear Programming, 2nd edn. (Springer, Berlin, 2003)
S.D. Müller, N. Hansen, P. Koumoutsakos, Increasing the serial and the parallel performance of the CMA-evolution strategy with large populations, in Proceedings of the 7th International Conference on Parallel Problem Solving from Nature (PPSN VII), Granada, ed. by J.J. Merelo et al. Volume 2439 of Lecture Notes in Computer Science (Springer, Berlin, 2002), pp. 422–431
A. Ostermeier, A. Gawelczyk, N. Hansen, A derandomized approach to self adaptation of evolution strategies. Evol. Comput. 2(4), 369–380 (1994)
A. Ostermeier, A. Gawelczyk, N. Hansen, Step-size adaptation based on non-local use of selection information, in Proceedings of the 3rd International Conference on Parallel Problem Solving from Nature (PPSN III), Jerusalem, ed. by Y. Davidor et al. Volume 866 of Lecture Notes in Computer Science (Springer, Berlin, 1994), pp. 189–198
I. Rechenberg, Evolutionsstrategie: Optimierung Technischer Systeme nach Prinzipien der biologischen Evolution (Frommann-Holzboog, Stuttgart, 1973)
I. Rechenberg, Evolutionsstrategie’94 (Frommann-Holzboog, Stuttgart, 1994)
R. Ros, N. Hansen, A simple modification in CMA-ES achieving linear time and space complexity, in Proceedings of the 10th International Conference on Parallel Problem Solving from Nature (PPSN X), Dortmund, ed. by G. Rudolph et al. Volume 5199 of Lecture Notes in Computer Science (Springer, Berlin, 2008), pp. 296–305
G. Rudolph, On correlated mutations in evolution strategies, in Proceedings of the 2nd International Conference on Parallel Problem Solving from Nature (PPSN II), Brussels, ed. by R. Männer, B. Manderick (Elsevier, Amsterdam, 1992), pp. 105–114
G. Rudolph, An evolutionary algorithm for integer programming, in Proceedings of the 3rd Conference on Parallel Problem Solving from Nature (PPSN III), Jerusalem, ed. by Y. Davidor et al. Volume 866 of Lecture Notes in Computer Science (Springer, Berlin, 1994), pp. 63–66
G. Rudolph, Convergence Properties of Evolutionary Algorithms (Kovač, Hamburg, 1997)
G. Rudolph, Evolutionary strategies, in Handbook of Natural Computing, ed. by G. Rozenberg, T. Bäck, J.N. Kok (Springer, Berlin, 2012)
H.-P. Schwefel, Kybernetische Evolution als Strategie der experimentellen Forschung in der Strömungstechnik. Diplomarbeit, Technische Universität Berlin, Hermann Föttinger–Institut für Strömungstechnik, 1964
H.-P. Schwefel, Numerische Optimierung von Computer-Modellen Mittels der Evolutionsstrategie (Birkhäuser, Basel, 1977)
H.-P. Schwefel, Numerical Optimization of Computer Models (Wiley, Chichester, 1981)
O.M. Shir, Niching in Derandomized Evolution Strategies and its Applications in Quantum Control. PhD thesis, University of Leiden, The Netherlands, 2008
A. Stuart, K. Ord, S. Arnold, Kendall’s Advanced Theory of Statistics, Classical Inference and the Linear Model. Volume 2 in Kendall’s Library of Statistics (Wiley, Chichester, 2009)
Y. Sun, D. Wierstra, T. Schaul, J. Schmidhuber, Efficient natural evolution strategies, in Proceedings of the 11th Annual Conference on Genetic and Evolutionary Computation (GECCO’09), Shanghai, ed. by F. Rothlauf et al. (ACM, New York, 2009), pp. 539–546
Y. Sun, D. Wierstra, T. Schaul, J. Schmidhuber, Stochastic search using the natural gradient, in Proceedings of the 26th Annual International Conference on Machine Learning, ICML’09, Montreal, ed. by A. Pohoreckyj Danyluk et al. (ACM, New York, 2009), pp. 1161–1168
B. Tang, Orthogonal array-based latin hypercubes. J. Am. Stat. Assoc. 88(424), 1392–1397 (1993)
S. Wessing, M. Preuss, G. Rudolph, When parameter tuning actually is parameter control, in Proceedings of the 13th Annual Conference on Genetic and Evolutionary Computation (GECCO’11), Dublin, ed. by N. Krasnogor et al. (ACM, New York, 2011), pp. 821–828
D. Wierstra, T. Schaul, J. Peters, J. Schmidhuber, Natural evolution strategies, in Proceedings of the IEEE Congress on Evolutionary Computation (CEC’08), Hong Kong (IEEE, Piscataway, 2008), pp. 3381–3387
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Bäck, T., Foussette, C., Krause, P. (2013). Evolution Strategies. In: Contemporary Evolution Strategies. Natural Computing Series. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-40137-4_2
Download citation
DOI: https://doi.org/10.1007/978-3-642-40137-4_2
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-40136-7
Online ISBN: 978-3-642-40137-4
eBook Packages: Computer ScienceComputer Science (R0)