基于受控旋转门的量子神经网络模型算法及应用

Controlled-rotating gate-based quantum neural networks model algorithm with application

提出一种量子神经网络模型及算法.首先借鉴受控非门的含义提出一种受控量子旋转门,基于该门的物理意义,提出一种量子神经元模型,该模型包含对输入量子比特相位的旋转角度和对旋转角度的控制量两种设计参数;然后基于上述量子神经元提出一种量子神经网络模型,基于梯度下降法详细设计了该模型的学习算法;最后通过模式识别和时间序列预测两个仿真验证了该模型及算法在收敛能力和鲁棒性方面优于普通的BP网络.

A quantum neural networks model and its learning algorithm are presented. Firstly, a quantum controlled-rotating gate is proposed by analyzing the meaning of the controlled-NOT gate. Then a quantum neuron model is constructed from the physical meaning of the controlled rotating gate. The model includes two kinds of design parameters, rotation angle and its control range. Secondly, a quantum neural networks model based on quantum neuron is proposed. By using gradient descent algorithm, a learning algorithm of this model is designed in detail. With applications of pattern recognition and time series prediction, the simulation results show that the proposed algorithm is superior to the common BP neural networks in both convergence capability and robustness.