Sciweavers

Share
EMO
2006
Springer

Prediction-Based Population Re-initialization for Evolutionary Dynamic Multi-objective Optimization

8 years 7 months ago
Prediction-Based Population Re-initialization for Evolutionary Dynamic Multi-objective Optimization
Abstract. Optimization in changing environment is a challenging task, especially when multiple objectives are to be optimized simultaneously. The basic idea to address dynamic optimization problems is to utilize history information to guide future search. In this paper, two strategies for population re-initialization are introduced when a change in the environment is detected. The first strategy is to predict the new location of individuals from the location changes that have occurred in the history. The current population is then partially or completely replaced by the new individuals generated based on prediction. The second strategy is to perturb the current population with a Gaussian noise whose variance is estimated according to previous changes. The prediction based population reinitialization strategies, together with the random re-initialization method, are then compared on two bi-objective test problems. Conclusions on the different re-initialization strategies are drawn based...
Aimin Zhou, Yaochu Jin, Qingfu Zhang, Bernhard Sen
Added 22 Aug 2010
Updated 22 Aug 2010
Type Conference
Year 2006
Where EMO
Authors Aimin Zhou, Yaochu Jin, Qingfu Zhang, Bernhard Sendhoff, Edward P. K. Tsang
Comments (0)
books