Evolutionary Optimization with Dynamic Fidelity Computational Models

Lim, Dudy; Ong, Yew-Soon; Jin, Yaochu; Sendhoff, Bernhard

doi:10.1007/978-3-540-85984-0_29

Cited by 23 publications

(18 citation statements)

References 20 publications

(30 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lim et al [32] use a lower fidelity model to locally optimize each solution in the population. The fidelity used is decided based on a user defined threshold η that specifies the minimum correlation required between the model chosen and the highest fidelity model.…”

Section: A Evolutionary Optimization Using Surrogate Modelsmentioning

confidence: 99%

Efficient Use of Partially Converged Simulations in Evolutionary Optimization

Branke

Asafuddoula

Bhattacharjee

et al. 2017

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Section: A Evolutionary Optimization Using Surrogate Modelsmentioning

confidence: 99%

Efficient Use of Partially Converged Simulations in Evolutionary Optimization

Branke

Asafuddoula

Bhattacharjee

et al. 2017

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

“…The best performing individuals from the bottom layer would then migrate into a top layer for more exact evaluation, and vice versa. Also in the class, Lim et al [23] explored an evolutionary based optimization algorithm which adjusted the simulation fidelity during the search, and based on the correlation between adjacent members in the population of candidate solutions.…”

Section: Expensive Optimization Problemsmentioning

confidence: 99%

“…Φi,j = φj(x x xi)(23) x x xi, i = 1, • • • , n, are the sample vectors, and f f f is the vector of corresponding objective function values. As mentioned in Section 3, the proposed algorithm used an radial basis functions network withn = 0.8 n, i.e., the number of neurons was 80 % of the sample size.…”

mentioning

confidence: 99%

An Optimization Algorithm Employing Multiple Metamodels and Optimizers

Tenne

2013

Int. J. Autom. Comput.

View full text Add to dashboard Cite

Modern engineering design optimization often relies on computer simulations to evaluate candidate designs, a setup which results in expensive black-box optimization problems. Such problems introduce unique challenges, which has motivated the application of metamodel-assisted computational intelligence algorithms to solve them. Such algorithms combine a computational intelligence optimizer which employs a population of candidate solutions, with a metamodel which is a computationally cheaper approximation of the expensive computer simulation. However, although a variety of metamodels and optimizers have been proposed, the optimal types to employ are problem dependant. Therefore, a priori prescribing the type of metamodel and optimizer to be used may degrade its effectiveness. Leveraging on this issue, this study proposes a new computational intelligence algorithm which autonomously adapts the type of the metamodel and optimizer during the search by selecting the most suitable types out of a family of candidates at each stage. Performance analysis using a set of test functions demonstrates the effectiveness of the proposed algorithm, and highlights the merit of the proposed adaptation approach.

show abstract

“…To enhance the accuracies of the surrogates used, researchers have turned to localized , Emmerich et al, 2002, Ong et al, 2003, Regis and Shoemaker, 2004 as opposed to globalized models or their synergies [Zhou et al, 2005]. Others have also considered the use of gradient information [Ong et al, 2004] to enhance the prediction accuracy of the constructed surrogate models or physics-based models that are deem to be more trustworthy than the purely data-centric counterparts [Keane and Petruzzelli, 2000, Keane and Petruzzelli, 2000, Lim et al, 2008.…”

Section: Introductionmentioning

confidence: 99%

Evolution by Adapting Surrogates

Ong

Menzel

et al. 2013

Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

To deal with complex optimization problems plagued with computationally expensive fitness functions, the use of surrogates to replace the original functions within the evolutionary framework is becoming a common practice. However, the appropriate datacentric approximation methodology to use for the construction of surrogate model would depend largely on the nature of the problem of interest, which varies from fitness landscape and state of the evolutionary search, to the characteristics of search algorithm used. This has given rise to the plethora of surrogate-assisted evolutionary frameworks proposed in the literature with ad hoc approximation/surrogate modeling methodologies considered. Since prior knowledge on the suitability of the data centric approximation methodology to use in surrogate-assisted evolutionary optimization is typically unavailable beforehand, this paper presents a novel evolutionary framework with the evolvability learning of surrogates (EvoLS) operating on multiple diverse approximation methodologies in the search. Further, in contrast to the common use of fitness prediction error as a criterion for the selection of surrogates, the concept of evolvability to indicate the productivity or suitability of an approximation methodology that brings about fitness improvement in the evolutionary search is introduced as the basis for adaptation. The backbone of the proposed EvoLS is a statistical learning scheme to determine the evolvability of each approximation methodology while the search progresses online. For each individual solution, the most productive approximation methodology is inferred, that is, the method with highest evolvability measure. Fitness improving surrogates are subsequently constructed for use within a trust-region enabled local search strategy, leading to the self-configuration of a surrogate-assisted memetic algorithm for solving computationally expensive problems. A numerical study of EvoLS on commonly used benchmark problems and a real-world computationally expensive aerodynamic car rear design problem highlights the efficacy of the proposed EvoLS in attaining reliable, high quality, and efficient performance under a limited computational budget.

show abstract

Evolutionary Optimization with Dynamic Fidelity Computational Models

Cited by 23 publications

References 20 publications

Efficient Use of Partially Converged Simulations in Evolutionary Optimization

Efficient Use of Partially Converged Simulations in Evolutionary Optimization

An Optimization Algorithm Employing Multiple Metamodels and Optimizers

Evolution by Adapting Surrogates

Contact Info

Product

Resources

About