An effective cooperative coevolution framework integrating global and local search for large scale optimization problems

Cao, Zijian; Wang, Lei; Shi, Yuhui; Hei, Xinhong; Rong, Xiaofeng; Jiang, Qiaoyong; Li, Hongye

doi:10.1109/cec.2015.7257129

Cited by 11 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…f) Cooperative Coevolution and Memetic Algorithms: Decomposition methods (see §II-B) and memetic algorithms are the two most widely used approaches to large-scale global optimization with algorithms from both categories ranked first in large-scale global optimization competitions [169,226,227]. To benefit from the advantages of both approaches, some authors suggest the use of memetic algorithms [125,126,191,207,213] or other hybrids [218] as component optimizers in a cooperative coevolution framework. The general approach is to decompose the problem into a set of lower dimensional subproblems using the methods described in §II-B, and optimize each component using a global search algorithm followed by an episode of local search.…”

Section: B) Choice Of Solutionsmentioning

confidence: 99%

“…The general approach is to decompose the problem into a set of lower dimensional subproblems using the methods described in §II-B, and optimize each component using a global search algorithm followed by an episode of local search. Cao et al [207] proposed to use SaNSDE [228] followed by Solis and Wets' [222] on each component and adjust their search intensity/frequency according to their performance. Sun et al [213] also used SaNSDE as the global search operator followed by dedicated local search procedures for the separable and nonseparable components.…”

Section: B) Choice Of Solutionsmentioning

confidence: 99%

See 1 more Smart Citation

A Review of Population-Based Metaheuristics for Large-Scale Black-Box Global Optimization—Part I

Omidvar

Yao

2022

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Scalability of optimization algorithms is a major challenge in coping with the ever growing size of optimization problems in a wide range of application areas from highdimensional machine learning to complex large-scale engineering problems. The field of large-scale global optimization is concerned with improving the scalability of global optimization algorithms, particularly population-based metaheuristics. Such metaheuristics have been successfully applied to continuous, discrete, or combinatorial problems ranging from several thousand dimensions to billions of decision variables. In this two-part survey, we review recent studies in the field of large-scale black-box global optimization to help researchers and practitioners gain a bird's-eye view of the field, learn about its major trends, and the state-of-the-art algorithms. Part A of the series covers two major algorithmic approaches to large-scale global optimization: problem decomposition and memetic algorithms. Part B of the series covers a range of other algorithmic approaches to largescale global optimization, describes a wide range of problem areas, and finally touches upon the pitfalls and challenges of current research and identifies several potential areas for future research.

show abstract

Section: B) Choice Of Solutionsmentioning

confidence: 99%

Section: B) Choice Of Solutionsmentioning

confidence: 99%

A Review of Population-Based Metaheuristics for Large-Scale Black-Box Global Optimization—Part I

Omidvar

Yao

2022

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

show abstract

“…The CCEA framework is the decomposition approach. There are three main decomposition approaches, including random grouping [37] , static grouping [38] , and grouping based on variable correlation [39,40] . The advantage of random grouping over static grouping is that the random grouping is more likely to optimize the interactive variables together [7] .…”

Section: Cooperative Co-evolutionary Algorithmmentioning

confidence: 99%

Large-Scale Expensive Optimization with a Switching Strategy

Sun

et al. 2022

Complex Syst. Model. Simul.

View full text Add to dashboard Cite

Some optimization problems in scientific research, such as the robustness optimization for the Internet of Things and the neural architecture search, are large-scale in decision space and expensive for objective evaluation. In order to get a good solution in a limited budget for the large-scale expensive optimization, a random grouping strategy is adopted to divide the problem into some low-dimensional subproblems. A surrogate model is then trained for each sub-problem using different strategies to select training data adaptively. After that, a dynamic infill criterion is proposed corresponding to the models currently used in the surrogate-assisted sub-problem optimization. Furthermore, an escape mechanism is proposed to keep the diversity of the population. The performance of the method is evaluated on CEC'2013 benchmark functions.Experimental results show that the algorithm has better performance in solving expensive large-scale optimization problems.

show abstract

“…In science and engineering, there are cases in which a search for the optimal solution in a large and complex space is required [1]. Traditional optimization algorithms, such as Newton's method and the gradient descent method [2], can solve the simple and continuous differentiable function [3].…”

Section: Introductionmentioning

confidence: 99%

Artificial Flora (AF) Optimization Algorithm

Cheng

Yan

2018

Applied Sciences

View full text Add to dashboard Cite

Featured Application: The proposed algorithm can be used in unconstrained multivariate function optimization problems, multi-objective optimization problems and combinatorial optimization problems.Abstract: Inspired by the process of migration and reproduction of flora, this paper proposes a novel artificial flora (AF) algorithm. This algorithm can be used to solve some complex, non-linear, discrete optimization problems. Although a plant cannot move, it can spread seeds within a certain range to let offspring to find the most suitable environment. The stochastic process is easy to copy, and the spreading space is vast; therefore, it is suitable for applying in intelligent optimization algorithm. First, the algorithm randomly generates the original plant, including its position and the propagation distance. Then, the position and the propagation distance of the original plant as parameters are substituted in the propagation function to generate offspring plants. Finally, the optimal offspring is selected as a new original plant through the selection function. The previous original plant becomes the former plant. The iteration continues until we find out optimal solution. In this paper, six classical evaluation functions are used as the benchmark functions. The simulation results show that proposed algorithm has high accuracy and stability compared with the classical particle swarm optimization and artificial bee colony algorithm.

show abstract

An effective cooperative coevolution framework integrating global and local search for large scale optimization problems

Cited by 11 publications

References 13 publications

A Review of Population-Based Metaheuristics for Large-Scale Black-Box Global Optimization—Part I

A Review of Population-Based Metaheuristics for Large-Scale Black-Box Global Optimization—Part I

Large-Scale Expensive Optimization with a Switching Strategy

Artificial Flora (AF) Optimization Algorithm

Contact Info

Product

Resources

About