.png){kind=icon}
返回
A scalable memory-enhanced swarm intelligence optimization method: fractional-order Bat-inspired algorithm
.png){kind=icon}
PRE
AI
DOI：10.1007/s13042-023-02022-1.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The Bat-inspired algorithm (BA), as one of the swarm intelligence algorithms, has a high potential for solving global optimization problems. This algorithm possesses an additional inherent capability compared to other swarm intelligence algorithms, which is the inclusion of a local search mechanism. Although this increases the convergence speed of the algorithm, excessive focus on exploitation, especially in the initial iterations, may lead to premature convergence and stagnation at the local optimum. In this paper, to address this drawback and strengthen the exploration capability while also achieving a balance between it and the exploitation capability, an improved version of BA, called fractional-order BA (FOBA), is introduced. The development of velocity and position vectors in FOBA, using the concept of fractional-order derivatives, extends the memory related to the previous behaviors of artificial bats and controls the convergence of the algorithm. To evaluate the proposed algorithm, ten well-known benchmark functions are used, and the results are compared with standard and state-of-the-art metaheuristic algorithms that have been introduced recently. Experimental results show that FOBA performs better than all compared algorithms. Furthermore, the proposed algorithm is used to optimize the weights and bias of the MLP neural network using six classification datasets. The results demonstrate the effectiveness and efficiency of FOBA.
Keyword:
Fractional calculus
Bat-inspired algorithm
Swarm intelligence
Exploration
Exploitation
Multi-layer perceptron
期刊
IF:
2.7
论文数:
3.2K
被引数:
5.6K
机构
引用论文
Complex Grunwald-Letnikov, Liouville, Riemann-Liouville, and Caputo derivatives for analytic functions解析函数的复杂grunwald-letnikov，Liouville，riemann-liouville和caput0导数
PREAI
分享收藏Bat algorithm based control to decrease the control energy consumption and modified bat algorithm based control to increase the trajectory tracking accuracy in robots.png)
NEURAL NETWORKS
NEURAL NETWORKSIF6.3
PREAI
分享收藏
Bat-inspired algorithms with natural selection mechanisms for global optimizationNEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏
查看更多内容