摘要
QDPSO(QuantumDelta-Potential-Well-basedParticleSwarmOptimization)算法是基于量子空间的粒子群算法,对QDPSO算法进行了改进,结合Iris分类问题,应用到BP网络的权值优化中,并和基于标准PSO算法的方法进行了比较。实验结果表明:该算法性能优于所比较的两种算法,并且具有良好的收敛性和稳定性。
QDPSO is the particle swarm optimization in quantum space. It was improved and applied in BP neural network training combining with Iris-classify problem. The proposed algorithm compared with that of which was based on the standard PSO. The results show that the proposed algorithm is superior to the other two algorithms with a better astringency and stability.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2005年第9期2078-2081,共4页
Journal of System Simulation
基金
国家自然科学基金(60474030)
关键词
神经网络
PSO算法
QDPSO算法
优化
Neural network
Particle swarm optimization
Quantum Delta-Potential-Well-based Particle swarm
optimization