-
为提高过程神经网络的逼近和泛化能力, 从研究过程神经元信息处理的量子计算实现机理入手, 提出基于量子旋转门及多位受控非门的物理意义构造量子过程神经元的新思想. 将离散化后的过程式输入信息作为受控非门的控制位, 经过量子旋转门作用后控制目标量子位的状态, 以目标量子位处于状态|1概率幅作为量子过程神经元的输出. 以量子过程神经元为隐层, 普通神经元为输出层, 可构成量子过程神经网络. 基于量子计算机理推导了该模型的学习算法. 将该模型用于太阳黑子数年均值预测, 应用结果表明, 所提方法与普通过程神经网络相比, 预测精度有所提高, 对于复杂预测问题具有一定理论意义和实用价值.To enhance the approximation and generalization ability of process neural networks (PNNs), by studying the quantum implementation mechanism of information processing of process neuron, a new idea of designing quantum process neuron is proposed in this paper, based on the quantum rotation gates and the quantum controlled-non gates. In the proposed approach, the discrete process inputs are expressed by the qubits, which, as the control qubits of controlled-non gates after being rotated by the quantum rotation gates, control the target qubits to reverse. The model outputs are described by the probability amplitude of state |1 in the target qubits. Then the quantum process neural networks (QPNNs) are designed by the quantum process neurons for the hidden layer and the normal neurons for the output layer. The algorithm of QPNN is derived through the quantum computing. The proposed approach is utilized to predict the smoothed yearly mean sunspot numbers, and the results indicate that the QPNN has higher prediction accuracy than the normal PNN, thus it has a certain theoretical meaning and practical value for the complex prediction.
-
Keywords:
- quantum computation /
- quantum process neuron /
- quantum process neural networks /
- algorithm design
[1] Tsoi A C 1994 IEEE Trans. Neural Networ. 7 229
[2] He X G, Liang J Z 2000 Proceedings of the 16th World Computer Conferences on Intelligent Information Processing Beijing, China, August 12-15, 2000 p143
[3] He X G, Liang J Z 2000 Eng. Sci. 2 40 (in Chinese) [何新贵, 梁久祯 2000 中国工程科学 2 40]
[4] He X G, Liang J Z, Xu S H 2001 Eng. Sci. 3 31 (in Chinese) [何新贵, 梁久祯, 许少华 2001 中国工程科学 3 31]
[5] Xu S H, He X G, Liu K 2006 Acta Electron. Sin. 34 1838 (in Chinese) [许少华, 何新贵, 刘坤 2006 电子学报 34 1838]
[6] Xu S H, He X G, Li P C 2003 J. Comput. Res. Dev. 40 1612 (in Chinese) [许少华, 何新贵, 李盼池 2003 计算机研究与发展 40 1612]
[7] Xu S H, He X G 2004 J. BUAA 30 14 (in Chinese) [许少华, 何新贵 2004 北京航空航天大学学报 30 14]
[8] Xu S H, He X G 2004 Pattern Recogn. Artif. Intell. 17 201 (in Chinese) [许少华, 何新贵 2004 模式识别与人工智能 17 201]
[9] Zhong S S, Ding G 2005 Control Decis. 20 764 (in Chinese) [钟诗胜, 丁刚 2005 控制与决策 20 764]
[10] Liang J Z 2006 Pattern Recogn. Artif. Intell. 19 295 (in Chinese) [梁久祯 2006 模式识别与人工智能 19 295]
[11] He X G, Xu S H 2004 Acta Automatica Sin. 30 801 (in Chinese) [何新贵, 许少华 2004 自动化学报 30 801]
[12] Xu S H, Li P C, He X G 2009 CAAI Trans. Intell. Syst. 4 283 (in Chinese) [许少华, 李盼池, 何新贵 2009 智能系统学报 4 283]
[13] Ding G, Zhong S S 2007 Acta Phys. Sin. 56 1224 (in Chinese) [丁刚, 钟诗胜 2007 56 1224]
[14] Michiharu M, Masaya S, Hiromi M 2007 Appl. Math. Comput. 185 1015
[15] Li P C, Song K P, Yang E L 2010 Neural Netw. World 20 189
-
[1] Tsoi A C 1994 IEEE Trans. Neural Networ. 7 229
[2] He X G, Liang J Z 2000 Proceedings of the 16th World Computer Conferences on Intelligent Information Processing Beijing, China, August 12-15, 2000 p143
[3] He X G, Liang J Z 2000 Eng. Sci. 2 40 (in Chinese) [何新贵, 梁久祯 2000 中国工程科学 2 40]
[4] He X G, Liang J Z, Xu S H 2001 Eng. Sci. 3 31 (in Chinese) [何新贵, 梁久祯, 许少华 2001 中国工程科学 3 31]
[5] Xu S H, He X G, Liu K 2006 Acta Electron. Sin. 34 1838 (in Chinese) [许少华, 何新贵, 刘坤 2006 电子学报 34 1838]
[6] Xu S H, He X G, Li P C 2003 J. Comput. Res. Dev. 40 1612 (in Chinese) [许少华, 何新贵, 李盼池 2003 计算机研究与发展 40 1612]
[7] Xu S H, He X G 2004 J. BUAA 30 14 (in Chinese) [许少华, 何新贵 2004 北京航空航天大学学报 30 14]
[8] Xu S H, He X G 2004 Pattern Recogn. Artif. Intell. 17 201 (in Chinese) [许少华, 何新贵 2004 模式识别与人工智能 17 201]
[9] Zhong S S, Ding G 2005 Control Decis. 20 764 (in Chinese) [钟诗胜, 丁刚 2005 控制与决策 20 764]
[10] Liang J Z 2006 Pattern Recogn. Artif. Intell. 19 295 (in Chinese) [梁久祯 2006 模式识别与人工智能 19 295]
[11] He X G, Xu S H 2004 Acta Automatica Sin. 30 801 (in Chinese) [何新贵, 许少华 2004 自动化学报 30 801]
[12] Xu S H, Li P C, He X G 2009 CAAI Trans. Intell. Syst. 4 283 (in Chinese) [许少华, 李盼池, 何新贵 2009 智能系统学报 4 283]
[13] Ding G, Zhong S S 2007 Acta Phys. Sin. 56 1224 (in Chinese) [丁刚, 钟诗胜 2007 56 1224]
[14] Michiharu M, Masaya S, Hiromi M 2007 Appl. Math. Comput. 185 1015
[15] Li P C, Song K P, Yang E L 2010 Neural Netw. World 20 189
计量
- 文章访问数: 8236
- PDF下载量: 689
- 被引次数: 0
下载:
