一种基于ABC的量子神经网络训练算法

传统的量子神经网络的训练方法容易使得算法陷入局部极小值,将Artificial Bee Colony(ABC)算法引入到原训练算法中,并且对人工蜂群算法进行改进.利用改进后的人工蜂群算法来优化传统量子神经网络,使优化后的量子神经网络具有结构简单、参数少、收敛速度快和可跳出局部极小值等优点.实验结果表明,相比原训练算法该优化算法提高了量子神经网络收敛解的精度.

Artificial Bee Colony based training algorithm for Quantum Neural Networks

In order to solve the problem that the training result of Quantum Neural Network is easy to fall into local minimum, Artificial Bee Colony (ABC) algorithm is introduced into the original training algorithm to design an improved Artificial Bee Colony algorithm.Using the improved Artificial Bee Colony algorithm to optimize the traditional Quantum Neural Network, so that the optimized Quantum Neural Network has the advantages of simple structure, good generalization and being deviated from local minima.Experimental results show that compared with the original training algorithm, the optimization algorithm improves the convergence precision of Quantum Neural Network and accelerates the convergence speed of Quantum Neural Network.