.png){kind=icon}
.png){kind=icon}
Return
Surrogate-Assisted Evolutionary Framework with Adaptive Knowledge Transfer for Multi-Task Optimization
.png){kind=icon}
PRE
AI
DOI：10.1109/TETC.2019.2945775.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Multi-task optimization is a hot research topic in the field of evolutionary computation. This paper proposes an efficient surrogate-assisted multi-task evolutionary framework (named SaEF-AKT) with adaptive knowledge transfer for multi-task optimization. In the proposed SaEF-AKT, several tasks which are computationally expensive are solved jointly in each generation. Surrogate models are built based on the historical search information of each task to reduce the number of fitness evaluations. To improve the search efficiency, a general similarity measure mechanism and an adaptive knowledge transfer mechanism are proposed, which can help knowledge transfer among the tasks to be solved. The proposed SaEF-AKT is tested on a number of benchmark problems in multi-task optimization scenario and real-world time series regression problems. The experimental results demonstrate that the proposed framework can outperform several state-of-the-art multi-task optimization algorithms.
Keywords:
Task analysis
Optimization
Knowledge transfer
Evolutionary computation
Sociology
Statistics
Benchmark testing
Evolutionary multi-task
efficient global optimization
multi-problem surrogates
knowledge transfer
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
5.4
Papers:
1.1K
Citations:
3.4K