Research in the Performance Assessment of Multi-objective Optimization Evolutionary Algorithms
Guoqiang Deng, Zhangcan Huang, Min Tang · 2007
The use of evolutionary algorithms (EAs) for search and optimization tasks has become very popular in the last few years. Improving the existing algorithms or presenting new algorithms will necessarily refer to the performance assessment of these algorithms. Measuring the performance of algorithms has a basic issue: whether there exists a standard methodology that various multi-objective optimization evolutionary algorithms (MOEAs) can be directly compared. Unfortunately, researchers haven't paid much attention to this issue. This paper reviews some of the most representative assessment methodologies used in the literature and then provides some useful suggestions and advices for researchers of algorithms.