2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence) 2008
DOI: 10.1109/cec.2008.4631150
|View full text |Cite
|
Sign up to set email alerts
|

Approaches to selection and their effect on fitness modelling in an Estimation of Distribution Algorithm

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

2
38
0

Year Published

2008
2008
2021
2021

Publication Types

Select...
4
3

Relationship

2
5

Authors

Journals

citations
Cited by 26 publications
(40 citation statements)
references
References 19 publications
2
38
0
Order By: Relevance
“…The coefficients pointed towards application of the nematodes at points which matched the lifecycle of the fly larvae. Analysis of univariate and bivariate model coefficients for a number of further fitness functions was conducted in [4] and reinforced this finding.…”
Section: Model Quality and The Fitness Prediction Correlationmentioning
confidence: 78%
See 1 more Smart Citation
“…The coefficients pointed towards application of the nematodes at points which matched the lifecycle of the fly larvae. Analysis of univariate and bivariate model coefficients for a number of further fitness functions was conducted in [4] and reinforced this finding.…”
Section: Model Quality and The Fitness Prediction Correlationmentioning
confidence: 78%
“…While the number of function evaluations to reach an optimum were significantly fewer than with other algorithms in our earlier work the time taken to build and sample the probabilistic model was large. However, it has been found that the Markov network at the heart of DEUM represents a good model of fitness [5,7], and this has been exploited in recent work [8] as well as providing an explanation for the previous good results.…”
mentioning
confidence: 99%
“…This research trend has received increasing attention in the field of EDAs (Brownlee et al 2008;Sastry et al 2006;Shakya et al 2005).…”
Section: Function Approximation Modulementioning
confidence: 99%
“…To unveil and extract problem information that is hidden in the original formulation of the function or optimization problem (Brownlee et al 2008).…”
Section: Function Approximation Modulementioning
confidence: 99%
“…Fitness inheritance modeling has been incorporated into EcGA ) and BOA to estimate the fitness value of new individuals and, consequently, reduce the total number of function evaluations consumed by the algorithm to reach the optimum. Brownlee et al (2008) and Brownlee (2009) used a Markov network-based fitness model in DEUM (Shakya 2006) to predict the fitness value of the new individuals and also compare their correlations to the true fitness values.…”
Section: Other Modeling Typesmentioning
confidence: 99%