Journal
ARTIFICIAL INTELLIGENCE REVIEW
Volume 51, Issue 2, Pages 219-253Publisher
SPRINGER
DOI: 10.1007/s10462-017-9562-6
Keywords
Differential evolution; Learning-forgetting; Prior knowledge; Opposition learning
Categories
Funding
- National Natural Science Foundation of China [61603244]
Ask authors/readers for more resources
The search capability of differential evolution (DE) is largely affected by control parameters, mutation and crossover strategies. Therefore, choosing appropriate strategies and control parameters to solve different types of optimization problems or adapt distinct evolution phases is an important and challenging task. To achieve this objective, a DE with strategy adaptation and knowledge-based control parameters (SAKPDE) is proposed in the current study. In the proposed algorithm, a learning-forgetting mechanism is used to implement the adaptation of mutation and crossover strategies. Meanwhile, prior knowledge and opposition learning are utilized to supervise and guide the evolution of control parameters during the entire evolutionary process. SAKPDE is compared with eight improved DEs and four non-DE evolutionary algorithms using three well-known test suites (i.e., BBOB2012, IEEE CEC2005, and IEEE CEC2014). The results indicate that the average performance of SAKPDE is highly competitive among all compared algorithms.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available