Using semi-independent variables to enhance optimization search
作者：
Highlights：
• A heuristic, semi-independent variable formulation, is introduced to enhance optimization.
• This formulation embodies a set of expected or desired relationships among the original variables.
• Variable interaction analysis is used to measure the complexity of various formulations.
• The efficacy of the heuristic is demonstrated by testing it on various optimization algorithms.
• Population-based algorithms benefited the most from the proposed method.
摘要
•A heuristic, semi-independent variable formulation, is introduced to enhance optimization.•This formulation embodies a set of expected or desired relationships among the original variables.•Variable interaction analysis is used to measure the complexity of various formulations.•The efficacy of the heuristic is demonstrated by testing it on various optimization algorithms.•Population-based algorithms benefited the most from the proposed method.
论文关键词：Semi-independent variable,Heuristics,Evolutionary computation,Swarm intelligence
论文评审过程：Received 3 May 2018, Revised 23 November 2018, Accepted 24 November 2018, Available online 24 November 2018, Version of Record 3 December 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.11.031
