An experimental investigation of self-adaptation in evolutionary programming

Liang, Ko-Hsin; Yao, Xin; Liu, Yong; Newton, Charles S.; Hoffman, David

doi:10.1007/bfb0040782

Cited by 33 publications

(15 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus self-adaptation can be biased toward lowering the standard deviations too quickly resulting in stagnation (described as an extremely low rate of convergence). As a safeguard against such stagnation, a lower bound for the strategy parameters has been proposed [2], [6], used [4], [5], [7], and investigated [12]. When a lower bound is used, all 's that fall below in (2) are reset to .…”

Section: A Conventional Evolutionary Programmingmentioning

confidence: 99%

Combining mutation operators in evolutionary programming

Chellapilla

1998

IEEE Trans. Evol. Computat.

216

View full text Add to dashboard Cite

Abstract-Traditional investigations with evolutionary programming (EP) for continuous parameter optimization problems have used a single mutation operator with a parameterized probability density function (pdf), typically a Gaussian. Using a variety of mutation operators that can be combined during evolution to generate pdf's of varying shapes could hold the potential for producing better solutions with less computational effort. In view of this, a linear combination of Gaussian and Cauchy mutations is proposed. Simulations indicate that both the adaptive and nonadaptive versions of this operator are capable of producing solutions that are statistically as good as, or better, than those produced when using Gaussian or Cauchy mutations alone.

show abstract

Section: A Conventional Evolutionary Programmingmentioning

confidence: 99%

Combining mutation operators in evolutionary programming

Chellapilla

1998

IEEE Trans. Evol. Computat.

216

View full text Add to dashboard Cite

show abstract

“…Liang et al [43,44] considered the problem of a prematurely reduced mutation strength. They started with an empirical investigation on the loss of step size control for EP on five test functions [43].…”

Section: Problems and Limitations Of Self-adaptationmentioning

confidence: 99%

“…They started with an empirical investigation on the loss of step size control for EP on five test functions [43]. The EP used a population size of µ = 100 and a tournament size of q = 10.…”

Section: Problems and Limitations Of Self-adaptationmentioning

confidence: 99%

“…Stagnation of the search occurred even for the sphere model. As they argued [43,44], this might be due to the selection of an individual with a mutation strength far too small in one dimension but with a high fitness value. This individual propagates its ill-adapted mutation strength to all descendants and, therefore, the search stagnates.…”

Section: Problems and Limitations Of Self-adaptationmentioning

confidence: 99%

“…A way to circumvent such behavior is to introduce a lower bound for the step size. Fixed lower bounds are considered in [43]. While this surely prevents premature convergence of the EP, it does not take into account the fact that the ideal lower bound of the mutation strength depends on the actual state of the search.…”

Section: Problems and Limitations Of Self-adaptationmentioning

confidence: 99%

See 2 more Smart Citations

Self-Adaptation in Evolutionary Algorithms

Meyer-Nieberg

Beyer

Studies in Computational Intelligence

142

View full text Add to dashboard Cite

Summary. In this chapter, we will give an overview over self-adaptive methods in evolutionary algorithms. Self-adaptation in its purest meaning is a state-of-the-art method to adjust the setting of control parameters. It is called self-adaptive because the algorithm controls the setting of these parameters itself -embedding them into an individual's genome and evolving them. We will start with a short history of adaptation methods. The section is followed by a presentation of classification schemes for adaptation rules. Afterwards, we will review empirical and theoretical research of self-adaptation methods applied in genetic algorithms, evolutionary programming, and evolution strategies.

show abstract

Evolutionary Computation

Yao¹

International Series in Operations Research &Amp; Management Science

View full text Add to dashboard Cite

An experimental investigation of self-adaptation in evolutionary programming

Cited by 33 publications

References 4 publications

Combining mutation operators in evolutionary programming

Combining mutation operators in evolutionary programming

Self-Adaptation in Evolutionary Algorithms

Evolutionary Computation

Contact Info

Product

Resources

About