.png){kind=icon}
.png){kind=icon}
Return
A resource allocation-based multi-objective evolutionary algorithm for large-scale multi-objective optimization
.png){kind=icon}
PRE
AI
DOI：10.1007/s00500-023-09061-4.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In large-scale multi-objective optimization problems (LSMOPs), multiple conflicting objectives and hundreds even thousands of decision variables are contained. Therefore, it is a great challenge to address LSMOPs due to the curse of dimensionality. To tackle LSMOPs, this paper proposes a resource allocation-based multi-objective optimization evolutionary algorithm. In the proposed algorithm, decision variables are firstly divided into convergence-related variables and diversity-related variables by the proposed layer thickness-based variable classification (LTVC) method. Then, a resource allocation-based convergence optimization strategy is introduced for the convergence-related variables, which can allocate more computational resource to the sub-component with the best contribution. For the diversity-related variables, diversity optimization technique is adopted. Finally, the experimental results verify that the proposed algorithm has a competitive performance compared with some state-of-the-art algorithms.
Keywords:
Decision variable classification
Resource allocation
Large-scale optimization
Evolutionary multi-objective optimization