.png){kind=icon}
Return
MBCMEA: multi-armed bandit model-based constraint multimodal multi-objective optimization algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2025.129359.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A predictor based on a multi-armed bandit model is proposed. By solving this model, the value of using historical information, real-time data, and future prediction information in the current evolutionary phase of the population is assessed, thereby enabling the rational utilization of different population information. • A multi-population information hybrid evolution mechanism is proposed, where multiple populations evolve based on different population dynamics, helping the algorithm to expand the local search range. Additionally, a multi-stage evolution strategy is introduced to reasonably control the timing of information exchange between populations, thereby improving the utilization of population information. • The algorithm has been tested on multiple benchmark datasets and real-world problem, along with several strategy validation experiments. The experimental results demonstrate that the proposed algorithm is competitive.
Keywords:
multi-armed bandit
multi-population evolution
information utilization
evolutionary algorithm
hybrid strategy
Journal
IF:
7.5
Papers:
3.0W
Citations:
10.2W
Organization
Cited Papers
No cited papers available