Sizing the population with respect to the local progress in (1,λ)-evolution strategies - a theoretical analysis

Nikolaus Hansen; Andreas Gawelczyk; Andreas Ostermeier

Sizing the population with respect to the local progress in (1,λ)-evolution strategies - a theoretical analysis

Nikolaus Hansen
, Andreas Gawelczyk
, Andreas Ostermeier

TU Berlin

Research output: Contribution to conference › Paper › peer-review

24 Citations (Scopus)

Abstract

This paper presents an analysis of the local serial rate of progress with respect to the number of offspring λ for the (1,λ)-evolution strategy. It is shown that local serial progress is maximized when the expected progress of the second best offspring is zero. The theoretical results lead to a simple but efficient adaptation rule for λ, which needs no extra fitness function evaluations and only small computational expense. Simulations of the λ-adaptation on simple test functions are shown.

Original language	English
Pages	80-85
Number of pages	6
Publication status	Published - 1 Dec 1995
Externally published	Yes
Event	Proceedings of the 1995 IEEE International Conference on Evolutionary Computation. Part 1 (of 2) - Perth, Aust Duration: 29 Nov 1995 → 1 Dec 1995

Conference

Conference	Proceedings of the 1995 IEEE International Conference on Evolutionary Computation. Part 1 (of 2)
City	Perth, Aust
Period	29/11/95 → 1/12/95

Cite this

@conference{fb7acb0119514be687cb088aecf07831,

title = "Sizing the population with respect to the local progress in (1,λ)-evolution strategies - a theoretical analysis",

abstract = "This paper presents an analysis of the local serial rate of progress with respect to the number of offspring λ for the (1,λ)-evolution strategy. It is shown that local serial progress is maximized when the expected progress of the second best offspring is zero. The theoretical results lead to a simple but efficient adaptation rule for λ, which needs no extra fitness function evaluations and only small computational expense. Simulations of the λ-adaptation on simple test functions are shown.",

author = "Nikolaus Hansen and Andreas Gawelczyk and Andreas Ostermeier",

year = "1995",

month = dec,

day = "1",

language = "English",

pages = "80--85",

note = "Proceedings of the 1995 IEEE International Conference on Evolutionary Computation. Part 1 (of 2) ; Conference date: 29-11-1995 Through 01-12-1995",

}

Sizing the population with respect to the local progress in (1,λ)-evolution strategies - a theoretical analysis. / Hansen, Nikolaus; Gawelczyk, Andreas; Ostermeier, Andreas.
1995. 80-85 Paper presented at Proceedings of the 1995 IEEE International Conference on Evolutionary Computation. Part 1 (of 2), Perth, Aust.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Sizing the population with respect to the local progress in (1,λ)-evolution strategies - a theoretical analysis

AU - Hansen, Nikolaus

AU - Gawelczyk, Andreas

AU - Ostermeier, Andreas

PY - 1995/12/1

Y1 - 1995/12/1

N2 - This paper presents an analysis of the local serial rate of progress with respect to the number of offspring λ for the (1,λ)-evolution strategy. It is shown that local serial progress is maximized when the expected progress of the second best offspring is zero. The theoretical results lead to a simple but efficient adaptation rule for λ, which needs no extra fitness function evaluations and only small computational expense. Simulations of the λ-adaptation on simple test functions are shown.

AB - This paper presents an analysis of the local serial rate of progress with respect to the number of offspring λ for the (1,λ)-evolution strategy. It is shown that local serial progress is maximized when the expected progress of the second best offspring is zero. The theoretical results lead to a simple but efficient adaptation rule for λ, which needs no extra fitness function evaluations and only small computational expense. Simulations of the λ-adaptation on simple test functions are shown.

M3 - Paper

AN - SCOPUS:0029519739

SP - 80

EP - 85

T2 - Proceedings of the 1995 IEEE International Conference on Evolutionary Computation. Part 1 (of 2)

Y2 - 29 November 1995 through 1 December 1995

ER -

Sizing the population with respect to the local progress in (1,λ)-evolution strategies - a theoretical analysis

Abstract

Conference

Fingerprint

Cite this