A framework for evolutionary optimization with approximate fitness functions

By Yaochu Jin, Markus Olhofer, Bernhard Sendhoff. IEEE Transactions on Evolutionary Computation, 2002

692 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

View Paper: https://www.semanticscholar.org/paper/edab531e27934b7f10a0e7a7f78b2358f93374e9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

Citations in the corpus

Resampling Methods for Meta-Model Validation with Recommendations for Evolutionary Computation (2012, methodology): “In Lim et al. (2010) a hold-out approach is presented in which the test data is used for validating the prediction quality of the meta-model.”
Committee-Based Active Learning for Surrogate-Assisted Particle Swarm Optimization of Expensive Problems (2017, methodology): “ing to the surrogate [9], [21], because these solutions might improve the accuracy of the model in the promising region of search space and lead to better solutions on the basis of”
Review of surrogate modeling in water resources (2012, methodology): “…toward the already explored regions of the feasible space (probably containing local optima) and could be quite misleading; therefore, the unexplored regions would likely remain unexplored when optimizing on such metamodels [Jin et al., 2002; Regis and Shoemaker, 2007a; Yan and Minsker, 2006].”
Generalizing Surrogate-Assisted Evolutionary Computation (2010, background): “Such forms of model management schemes are termed as ‘evolution control’ in [7] and [20].”
ParEGO: a hybrid algorithm with on-line landscape approximation for expensive multiobjective optimization problems (2006, background): “Detailed reviews of all these meta-model based EAs can now be found in recent review papers [48], [ 34 ], [33], along with new approaches that try to manage the metamodels optimally with respect to the conicting concerns of optimization and the design of experiments (i.e.”
Single- and multiobjective evolutionary optimization assisted by Gaussian random field metamodels (2006, methodology): “EA using computationally expensive evaluations may reduce their CPU cost through metamodeling [2]‐[ 11 ].”
A Classification-Based Surrogate-Assisted Evolutionary Algorithm for Expensive Many-Objective Optimization (2019, methodology): “, the neural network assisted evolution strategy [32], the feasibility structure modeling assisted memetic algorithm [33], the classificationassisted memetic algorithm [34], and the surrogate-assisted cooperative particle swarm optimization [35].”
A Surrogate-Assisted Reference Vector Guided Evolutionary Algorithm for Computationally Expensive Many-Objective Optimization (2018, methodology): “In SAEAs, which individuals are to be re-evaluated using the original objective functions, how to update the surrogate and when to update the surrogate are referred to as model management, which is also known as evolution control [26].”
Surrogate-Assisted Cooperative Swarm Optimization of High-Dimensional Expensive Problems (2017, methodology): “The most commonly used surrogate models include polynomial regression [19], also known as response surface methodology [19], support vector machines [20]–[22], ANNs [12], [23], [24], radial basis function (RBF) networks [25]–[29], and Gaussian processes (GPs), also referred as to Kriging or design a”
Neuroevolutionary reinforcement learning for generalized control of simulated helicopters (2011, positive): “However, it is most commonly used to reduce computation costs [37, 59, 67, 68, 70] by ﬁnding a model that is faster than the original ﬁtness function.”
Numerical assessment of metamodelling strategies in computationally intensive optimization (2012, background): “Similarly, in the paper by Jin et al. (2002) which proposes the concept of evolution control in metamodel-enabled optimizers based on the metamodel embedded evolution approach (see Section 2.2.2), no attention/effort was devoted to avoid ANN over-training.”
On Using Surrogates with Genetic Programming (2015, methodology): “…generations that are evaluated with the surrogate model and generations that are evaluated using the expensive ﬁtness function (Rattle, 1998; Jin et al., 2002), to use surrogate models for supporting local search (Lim et al., 2010), or for pre-selection amongst individuals generated by a…”
Expensive Multiobjective Optimization by MOEA/D With Gaussian Process Model (2010, positive): “Much effort has been made to incorporate the Gaussian stochastic process and other response surface methods [4], [8], [9] into evolutionary algorithm (EA) frameworks for single objective optimization [10]–[17].”
Memetic algorithms and memetic computing optimization: A literature review (2012, background): “In [82] the Controlled Evaluations (CE) framework has been proposed.”