Abstract
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.
Highlights
The last few decades have seen tremendous progress in the field of evolutionary computation
Many experts and scholars have continuously devoted themselves to the study of evolutionary algorithms and have proposed numerous novel, well-established methods, such as the grey wolf optimizer (GWO) [5], ant lion optimizer (ALO) [6], teachinglearning-based optimization (TLBO) [7], virus colony search (VCS) [8], Harris hawks optimization (HHO) [9] and nuclear reaction optimization (NRO) [10]
The modifications proposed in the multi-leader search (MLS)-estimation of distribution algorithm (EDA) consist of four components: a) a weighted maximum likelihood estimation (MLE) method to improve the mean point quality, b) a distribution enhancement strategy (DES) to diversify the distribution scope, c) an MLS mechanism to eliminate stagnation, and d) tan eigen coordinate framework to modify the direction of evolution
Summary
The last few decades have seen tremendous progress in the field of evolutionary computation. Hybrid algorithms tend to have more parameters that require adjustment, which will reduce the parameter sensitivity of such an algorithm Both the deficiencies described above and the development prospects of EDAs in the field of evolutionary computation due to their model-based characteristics prompt us to propose new algorithms of this type. From CEC 2016 to CEC 2018, most of the top algorithms were based on the use of distribution models for sampling, such as UMOEAII [46] in 2016, EBOwithCMAR [47] in 2017, and HSES [48] in 2018 In contrast to these promising EDA-based variants, we propose a novel nonhybrid EDA variant using a multi-leader search (MLS) mechanism in an eigen coordinate system, called the MLS-EDA. The conclusions and prospects of this research are presented
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
