Anisotropic adaptive variance scaling for Gaussian estimation of distribution algorithm

Ren, Zhigang; Liang, Yongsheng; Wang, Lin; Zhang, Aimin; Pang, Bei; Li, Biying

doi:10.1016/j.knosys.2018.02.001

Cited by 19 publications

(8 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, the EDA usually requires a large population size to avoid stagnation [18]. However, our MLS-EDA can achieve promising performance with a relatively small population size, thus demonstrating the effectiveness of our modifications in enriching the population diversity.…”

Section: Comparison With Several Well-established Methods On the Cmentioning

confidence: 77%

“…The MLS-EDA ranks in first place on the 30D and 50D tests with the smallest ranking value, while L-SHADE ranks best on the 100D tests. The last row of this table shows the ranking performances of the algorithms on all three test sets in terms of a synthetically calculated score, denoted by SR. SR is calculated in accordance with the rank values in the first three rows of Table 7, as shown in (18).…”

Section: Comparison With Six Modern Algorithms From Different Famimentioning

confidence: 99%

“…However, when solving multimodal problems, an EDA cannot effectively capture the features of such a problem through an estimated Gaussian distribution model. Moreover, the variances of the distribution in the slope directions may shrink rapidly in a later stage of optimization [18]. Thus, the traditional EDA tends to suffer from premature convergence and can easily become trapped in stagnation.…”

Section: Introductionmentioning

confidence: 99%

“…Further studies were conducted by Ren and his teammates [23], who expanded the AMS strategy to consider the mean quality. Moreover, they explored a novel AVS strategy with anisotropic adjustment based on the local fitness landscape [18]. Their other work was EDA 2 [24], in which an archive is adopted to store more promising solutions to revise the distribution scope.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An Estimation of Distribution Algorithm With Multi-Leader Search

2020

View full text Add to dashboard Cite

The estimation of distribution algorithm (EDA) is a well-known stochastic search method but is easily affected by the ill-shaped distribution of solutions and can thus become trapped in stagnation. In this paper, we propose a novel modified EDA with a multi-leader search (MLS) mechanism, namely, the MLS-EDA. To strengthen the exploration performance, an enhanced distribution model that considers the information of population and distribution is utilized to generate new candidates. Moreover, when the algorithm stagnates, the MLS mechanism will be activated to perform a local search and shrink the search scope. The performance of the MLS-EDA in addressing complex optimization problems is verified using the CEC 2014 and CEC 2017 testbeds with 30D, 50D and 100D tests. Several modern algorithms, including the top-performing methods in the CEC 2014 and CEC 2017 competitions, are considered as competitors. The competitive performance of our proposed MLS-EDA is discussed based on the comparison results.

show abstract

Section: Comparison With Several Well-established Methods On the Cmentioning

confidence: 77%

Section: Comparison With Six Modern Algorithms From Different Famimentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Estimation of Distribution Algorithm With Multi-Leader Search

2020

View full text Add to dashboard Cite

show abstract

“…Another GEDA variants explored by them is EDA 2 [16], using an archive that can save more promising solutions to adjust the evolution direction. Their recent research [17] also proposed a novel variances adjustment technique, which has the advantage of tuning the variances and main search direction of GEDA simultaneously.…”

Section: Introductionmentioning

confidence: 99%

A Gaussian Estimation of Distribution Algorithm With Random Walk Strategies and Its Application in Optimal Missile Guidance Handover for Multi-UCAV in Over-the-Horizon Air Combat

et al. 2019

View full text Add to dashboard Cite

To overcome the premature convergence caused by the ill-distribution of solutions in the basic Gaussian estimation of distribution algorithm (GEDA), this paper explores a novel GEDA variant with random walk strategies, namely RW-GEDA. In RW-GEDA, the weighted maximum likelihood estimation method is used to estimate the Gaussian distribution. The new candidates are sampled using a shifted mean to enhance exploration performance. When the algorithm stagnates, two random walk strategies, namely, Gaussian random walk and Lévy walk, are activated to enrich the population diversity. Moreover, RW-GEDA is executed in an Eigen coordinate framework to promote the evolution towards the dominant region. The performance of RW-GEDA is evaluated by using the CEC 2014 test suite and compared with other top algorithms from different communities as well as promising GEDA extensions. The statistical results demonstrate the competitive performance of our proposed RW-GEDA in terms of efficiency and accuracy. In addition, RW-GEDA is applied to solve the optimal missile guidance handover problem. To fill the gap in solving this problem, a novel missile guidance advantage model is established, and the optimal missile guidance handover is determined by optimizing the control variables of unmanned combat aerial vehicles. The validity and practicability of the problem model as well as the accuracy and efficiency of RW-GEDA are demonstrated by the experimental results. INDEX TERMS Gaussian estimation of distribution algorithm, CEC 2014, numerical optimization, air combat, UCAV.

show abstract

A novel ensemble estimation of distribution algorithm with distribution modification strategies

Wang¹,

Liang

et al. 2023

Complex Intell. Syst.

View full text Add to dashboard Cite

The canonical estimation of distribution algorithm (EDA) easily falls into a local optimum with an ill-shaped population distribution, which leads to weak convergence performance and less stability when solving global optimization problems. To overcome this defect, we explore a novel EDA variant with an ensemble of three distribution modification strategies, i.e., archive-based population updating (APU), multileader-based search diversification (MSD), and the triggered distribution shrinkage (TDS) strategy, named E3-EDA. The APU strategy utilizes historical population information to rebuild the search scope and avoid ill-shaped distributions. Moreover, it continuously updates the archive to avoid overfitting the distribution model. The MSD makes full use of the location differences among populations to evolve the sampling toward promising regions. TDS is triggered when the search stagnates, shrinking the distribution scope to achieve local exploitation. Additionally, the E3-EDA performance is evaluated using the CEC 2014 and CEC 2018 test suites on 10-dimensional, 30-dimensional, 50-dimensional and 100-dimensional problems. Moreover, several prominent EDA variants and other top methods from CEC competitions are comprehensively compared with the proposed method. The competitive performance of E3-EDA in solving complex problems is supported by the nonparametric test results.

show abstract

Anisotropic adaptive variance scaling for Gaussian estimation of distribution algorithm

Cited by 19 publications

References 25 publications

An Estimation of Distribution Algorithm With Multi-Leader Search

An Estimation of Distribution Algorithm With Multi-Leader Search

A Gaussian Estimation of Distribution Algorithm With Random Walk Strategies and Its Application in Optimal Missile Guidance Handover for Multi-UCAV in Over-the-Horizon Air Combat

A novel ensemble estimation of distribution algorithm with distribution modification strategies

Contact Info

Product

Resources

About