What is it about?
As we know, there are many multiobjective evolutionary algorithms (MOEAs) that change their behaviour when solving different problems. These differences in each problem cause the performance of MOEA to fluctuate, making it hard to develop new algorithms or apply existing ones to new problems. To deal with this, the community started working on automatically designing MOEAs based on existing components to create even more effective variants of the original MOEA.
Featured Image
Photo by Trevor Vannoy on Unsplash
Why is it important?
Analysing the impact of the choices of components is not common in the literature. Doing this is important because if we understand the choices of components we can develop new and better components.
Perspectives
Writing this article was a great pleasure as it has co-authors with whom I have had long-standing collaborations. This paper opens a path to understanding why the behaviour of algorithms changes in problems from the point of view of the components that are part of these meta-heuristics.
Yuri Lavinas
Tsukuba Daigaku
Read the Original
This page is a summary of: Component-wise analysis of automatically designed multiobjective algorithms on constrained problems, July 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3512290.3528719.
You can read the full text:
Resources
Contributors
The following have contributed to this page
