Multi-objective genetic local search algorithm using Kohonen’s neural map

Hakimi-Asiabar, Mehrdad; Ghodsypour, Seyed Hassan; Kerachian, Reza

doi:10.1016/j.cie.2008.10.010

Cited by 40 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A set of new chromosomes in the next generation of solutions is produced by learning with the SOM. Hakimi-Asiabar et al (2009) applied this approach to improve the genetic diversity of solutions. The SBMOGA features a grid of neurons that applies the concept of the learning rule with a SOM to improve local and global searches.…”

Section: The Self-organizing Map-based Multi-objective Genetic Algorithmmentioning

confidence: 99%

Application of non-animal–inspired evolutionary algorithms to reservoir operation: an overview

Jahandideh-Tehrani

Bozorg‐Haddad

Loáiciga

2019

Environ Monit Assess

View full text Add to dashboard Cite

Evolutionary algorithms (EAs) have been widely used to search for optimal strategies for the planning and management of water resources systems, particularly reservoir operation. This study provides a comprehensive diagnostic assessment of state of the art of the non-animal-inspired EA applications to reservoir optimization. This type of EAs does not mimic biologic traits and group strategies of animal (wild) species. A search of pertinent papers was applied to the journal citation reports (JCRs). A bibliometric survey identified 14 pertinent non-animal-inspired EAs, such as the genetic algorithm (GA), simulated annealing (SA), and differential evolution (DE) algorithms, most of which have a number of modified versions. The characteristics of non-animal-inspired EAs and their modified versions were discussed to identify the difference between EAs and how each EA was improved. Additionally, the type of application of non-animal-inspired EAs to different case studies was investigated, and comparisons were made between the performance of the applied EAs in the studied literature. The survey revealed that the GA is the most frequently applied algorithm, followed by the DE algorithm. Non-animal-inspired EAs are superior to the classical methods of reservoir optimization (e.g., the non-linear programming and dynamic programming) due to faster convergence, diverse solution space, and efficient objective function evaluation. Several non-animal-inspired EAs of recent vintage have been shown to outperform the classic GA, which was the first evolutionary algorithm applied to reservoir operation.

show abstract

Section: The Self-organizing Map-based Multi-objective Genetic Algorithmmentioning

confidence: 99%

Application of non-animal–inspired evolutionary algorithms to reservoir operation: an overview

Jahandideh-Tehrani

Bozorg‐Haddad

Loáiciga

2019

Environ Monit Assess

View full text Add to dashboard Cite

show abstract

“…Several conventional accelerating EC convergence methods have been proposed to solve this problem [12]. They include approximating the fitness landscape [15,16,22], adaptive evolution control methods [6], developing new mechanisms embedded into existent EC algorithms [14], information fusion methods [5,8,27] and so on.…”

Section: Conventional Techniquesmentioning

confidence: 99%

Accelerating IEC and EC searches with elite obtained by dimensionality reduction in regression spaces

Pei

Tokuda

2013

Evol. Intel.

View full text Add to dashboard Cite

We propose a method for accelerating interactive evolutionary computation (IEC) and evolutionary computation (EC) searches using elite obtained in one-dimensional spaces and use benchmark functions to evaluate the proposed method. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using interpolation or approximation method, finds the best coordinate from the approximated shape, obtains elite by combining the best n found coordinates, and uses the elite for the next generation of the IEC or EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and there is a higher possibility that the elite individual locates near the global optimum. We compare the proposal with the methods obtaining landscape in the original search space, and show our proposed method can save the computational time significantly. Experimental evaluations of the technique in differential evolution using a simulated IEC user (Gaussian mixture model with different dimensions) and 34 benchmark functions show that the proposed method accelerates IEC and EC search significantly. Response to Reviewers:English-native proofreader re-checked our manuscript throughly.

show abstract

“…For the recent research, there are two topics for ANN based ECs. Firstly, EC uses ANN as a direction tool to find the best search direction, which uses landscapes information of the EC search space to train the ANN and with this information the EC search direction is decided [24]. Secondly, EC uses the trained ANN, which confirms an input-output relationship for special application with domain knowledge.…”

Section: Hybrid With Other Approaches 1) Local Search Algorithmmentioning

confidence: 99%

“…For an instance, neural networks has the memorial function, and the genetic algorithm is a best search optimization tool. It might be good to fuse both techniques in such a way that the memorial function of neural networks is used to direct the search direction of genetic algorithms so that the genetic algorithm's convergence can be accelerated [24].…”

Section: Ec Fusion With Other Soft-computing Approachmentioning

confidence: 99%

A survey on accelerating evolutionary computation approaches

Pei

Tokuda

2011

2011 International Conference of Soft Computing and Pattern Recognition (SoCPaR)

View full text Add to dashboard Cite

In this paper, we review the research on acceleration convergence approaches of evolutionary computation (EC) and its concrete application in the academy and industry. Evolutionary computation uses iterative progress, which is often inspired by biological mechanisms of evolution, to solve the problems that are multi-modal, multi-objective, discontinuous, non-differential, noisy and not well-defined. In this survey, many acceleration approaches are summarized and clustered in recent two decades. Applications of the acceleration approaches are included. We propose three promising research directions and their concrete approaches. These include including search space landscape approximation, search space projection and search strategy study, and comprise the main further research directions to be implemented an efficient EC search. Finally, we discuss the future research on accelerating convergence approaches of EC, and motivate some new approaches.

show abstract

Multi-objective genetic local search algorithm using Kohonen’s neural map

Cited by 40 publications

References 23 publications

Application of non-animal–inspired evolutionary algorithms to reservoir operation: an overview

Application of non-animal–inspired evolutionary algorithms to reservoir operation: an overview

Accelerating IEC and EC searches with elite obtained by dimensionality reduction in regression spaces

A survey on accelerating evolutionary computation approaches

Contact Info

Product

Resources

About