Generalizing Surrogate-Assisted Evolutionary Computation

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

doi:10.1109/tevc.2009.2027359

Cited by 414 publications

(213 citation statements)

References 55 publications

Supporting

Mentioning

205

Contrasting

Unclassified

Order By: Relevance

“…As suggested in Lim et al [4], approximation errors introduced by surrogate models in the evolutionary algorithms can have both negative and positive impacts. The negative impact, called 'curse of uncertainty' indicates the phenomenon that inaccurate surrogates may lead to EA to a false optimum.…”

Section: Radial Basis Function Network (Rbfn)mentioning

confidence: 99%

“…Inspired by the idea in [4], in this paper, a two-layer surrogate model is proposed to assist the search in PSO. The surrogate model in the top layer serves to smooth out local optimums, thus speeding up the search.…”

Section: Radial Basis Function Network (Rbfn)mentioning

confidence: 99%

“…However, many engineering design optimization problems involve the use of high fidelity simulation methods such as Finite Element Analysis (FEA), Computational Fluid Dynamics(CFD) and Computational Electro Magnetics (CEM) for quality evaluations, which is often computationally expensive, ranging from several minutes to days of supercomputer time [4]. Since PSO typically requires thousands of evaluations to achieve an acceptable optimum solution, the application of PSO to this class of expensive problems becomes intractable.…”

Section: Introductionmentioning

confidence: 99%

“…One promising approach to reduce computation time for optimization of highly time-consuming optimization problems is to employ computationally cheap approximation models (surrogates) to replace in part the computationally expensive exact function evaluations. Over the recent years, surrogate model assisted evolutionary algorithms (SAEAs) have received increasing attention for addressing expensive optimization problems, because the computational effort required to build and use surrogates is usually much lower than that for expensive evaluations [4]. A variety of surrogate models (also called metamodels or approximation models) have been proposed to be used in EAs, such as Polynomial Regression (PR, also known as response surface method) [5], Artificial Neural Network (ANN) [6], Radial Basis Function (RBF) [7], and Gaussian Process (GP) (also referred to Kriging) [8].…”

Section: Introductionmentioning

confidence: 99%

“…Note that the global model is shared by all particles in the swarm, whilst individual local surrogates are built for different particles. Different to the EAs assisted by global and local surrogate models reported in [4,22,23], in our propose method, no local search is employed.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A two-layer surrogate-assisted particle swarm optimization algorithm

et al. 2014

View full text Add to dashboard Cite

Like most Evolutionary Algorithms (EAs), Particle Swarm Optimization (PSO) usually requires a large number of fitness evaluations to obtain a sufficiently good solution. This poses an obstacle for applying PSO to computationally expensive problems. This paper proposes a two-layer surrogateassisted PSO (TLSAPSO) algorithm, in which a global and a number of local surrogate models are employed for fitness approximation. The global surrogate model aims to smooth out the local optima of the original multimodal fitness function and guide the swarm to fly quickly to an optimum. In the meantime, a local surrogate model constructed using the data samples near the particle is built to achieve a fitness estimation as accurate as possible. The contribution of each surrogate in the search is empirically verified by experiments on uni-and multi-modal problems. The performance of the proposed TLSAPSO algorithm is examined on ten widely used benchmark problems, and the experimental results show that the proposed algorithm is effective and highly competitive with the state-of-the-art, especially for multimodal optimization problems.

show abstract

Section: Radial Basis Function Network (Rbfn)mentioning

confidence: 99%

Section: Radial Basis Function Network (Rbfn)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A two-layer surrogate-assisted particle swarm optimization algorithm

et al. 2014

View full text Add to dashboard Cite

show abstract

Surrogate-assisted multicriteria optimization: Complexities, prospective solutions, and business case

Allmendinger

Emmerich

Hakanen

et al. 2017

J. Multi-Crit. Decis. Anal.

View full text Add to dashboard Cite

Complexity in solving real-world multicriteria optimization problems often stems from the fact that complex, expensive, and/or time-consuming simulation tools or physical experiments are used to evaluate solutions to a problem. In such settings, it is common to use efficient computational models, often known as surrogates or metamodels, to approximate the outcome (objective or constraint function value) of a simulation or physical experiment. The presence of multiple objective functions poses an additional layer of complexity for surrogate-assisted optimization.For example, complexities may relate to the appropriate selection of metamodels for the individual objective functions, extensive training time of surrogate models, or the optimal use of many-core computers to approximate efficiently multiple objectives simultaneously. Thinking out of the box, complexity can also be shifted from approximating the individual objective functions to approximating the entire Pareto front. This leads to further complexities, namely, how to validate statistically and apply the techniques developed to real-world problems. In this paper, we discuss emerging complexity-related topics in surrogate-assisted multicriteria optimization that may not be prevalent in nonsurrogate-assisted single-objective optimization. These complexities are motivated using several real-world problems in which the authors were involved. We then discuss several promising future research directions and prospective solutions to tackle emerging complexities in surrogate-assisted multicriteria optimization. Finally, we provide insights from an industrial point of view into how surrogate-assisted multicriteria optimization techniques can be developed and applied within a collaborative business environment to tackle real-world problems.

show abstract

Use of Multifidelity and Surrogate Models in the Design and Development of Physics-Based Systems

et al. 2016

View full text Add to dashboard Cite

Building a complex system in a time and cost effective manner, that performs well, meets diverse user needs, and has no inimical emergent behaviors is a challenge. The solution is to model before building. End-to-end high fidelity (HF) computational modeling and simulation (M&S) of a complex system, while attractive, is often too time consuming and expensive and may preclude the use of traditional systems engineering (SE) development tools. Missing are practicable approaches to reduce this computational time and cost by segmenting M&S performance and behaviors, keeping the model chain coherent, and only modeling what is necessary. An efficient framework is presented that couples mixed fidelity and surrogate models to reduce M&S computational time and cost for analysis of component-level and systemlevel performance of complex physics-based problems. This framework is used in an example to modify a Nuclear Electromagnetic Pulse (NEMP) Bounded Wave Simulator (BWS). The framework is flexible and potentially extensible; it may provide systems engineers with a cost-effective alternative to current approaches for modeling physics-based engineering systems (e.g., radar, communications systems).

show abstract

Generalizing Surrogate-Assisted Evolutionary Computation

Cited by 414 publications

References 55 publications

A two-layer surrogate-assisted particle swarm optimization algorithm

A two-layer surrogate-assisted particle swarm optimization algorithm

Surrogate-assisted multicriteria optimization: Complexities, prospective solutions, and business case

Use of Multifidelity and Surrogate Models in the Design and Development of Physics-Based Systems

Contact Info

Product

Resources

About