.png){kind=icon}
Return
Self-adaptive resource allocation based on reinforcement learning for multi-concept multi-objective optimization
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2026.114851.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A metric for measuring the concept of advantage has been developed using reinforcement learning. The establishment of this metric refers to the convergence accuracy and evolutionary potential of concepts, which is conducive to more accurate and effective screening of favorable concepts. • A reinforcement learning based evolutionary resource allocation methodology has been proposed. This methodology can dynamically guide evolutionary resources to concepts with obvious advantages, while protecting potential concepts with future advantage potential from being overly eliminated. • A reinforcement learning based environment selection algorithm has been designed, which can reasonably select elite individuals from various concepts. It ensures that a sufficient number of advantageous concept individuals are selected, while retaining a certain number of potential advantageous concept individuals. And it abandon the concept of disadvantage that no longer has potential in advance.
Keywords:
reinforcement learning
multi-concept optimization
resource allocation
evolutionary algorithms
advantage metric
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W
Organization
Cited Papers
A Gentle Introduction to Reinforcement Learning and its Application in Different Fields.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAIShareSaveShareSave
View more