Research Article
Weights Optimization Method of Differential Evolution Based on Artificial Bee Colony Algorithm
@INPROCEEDINGS{10.1007/978-3-030-72792-5_49, author={Ying Wu and Zibo Qi and Ling Jiang and Zifeng Dai and Chen Zhang and Changsheng Zhang and Jian Xu}, title={Weights Optimization Method of Differential Evolution Based on Artificial Bee Colony Algorithm}, proceedings={Simulation Tools and Techniques. 12th EAI International Conference, SIMUtools 2020, Guiyang, China, August 28-29, 2020, Proceedings, Part I}, proceedings_a={SIMUTOOLS}, year={2021}, month={4}, keywords={Artificial bee colony algorithm Differential evolution Neural network Ensemble learning Deep learning}, doi={10.1007/978-3-030-72792-5_49} }- Ying Wu
Zibo Qi
Ling Jiang
Zifeng Dai
Chen Zhang
Changsheng Zhang
Jian Xu
Year: 2021
Weights Optimization Method of Differential Evolution Based on Artificial Bee Colony Algorithm
SIMUTOOLS
Springer
DOI: 10.1007/978-3-030-72792-5_49
Abstract
Differential evolution algorithm is a search and optimization strategy that simulates the process of biological evolution. In the initial stage of the algorithm, it is necessary to generate a series of deep neural networks with sufficient accuracy as the initial population of subsequent algorithms. In this article, an artificial bee colony search strategy is added to the cross-operation of the differential evolution algorithm to optimize the weight value. The artificial bee colony algorithm search operator is introduced to guide the search of the population to avoid individuals in the population from falling into a local optimal situation. The experiments in this article verify the validity of the method through the handwritten digit recognition data set. The final results show that in the process of obtaining the initial population, using the differential evolution weight optimization method of the artificial bee colony search strategy optimizes the process of the fitness calculation in the model. It significantly improves the accuracy of the first-generation population and speeds up the overall process of the algorithm.
