A fuzzy modeling method for complex systems is studied. The notation of general stochastic neural network (GSNN) is presented and a new modeling method is given based on the combination of the modified Takagi and Suge...A fuzzy modeling method for complex systems is studied. The notation of general stochastic neural network (GSNN) is presented and a new modeling method is given based on the combination of the modified Takagi and Sugeno's (MTS) fuzzy model and one-order GSNN. Using expectation-maximization(EM) algorithm, parameter estimation and model selection procedures are given. It avoids the shortcomings brought by other methods such as BP algorithm, when the number of parameters is large, BP algorithm is still difficult to apply directly without fine tuning and subjective tinkering. Finally, the simulated example demonstrates the effectiveness.展开更多
期望最大化贝努利高斯（BG）近似信息传递（EM-BG-AMP）算法中的BG模型因为具有对称性,在逼近实际信号先验分布时会受到限制;而期望最大化高斯混合近似信息传递（EM-GM-AMP）算法中的GM模型是BG模型的高阶形式,复杂度较高。为了解决以...期望最大化贝努利高斯（BG）近似信息传递（EM-BG-AMP）算法中的BG模型因为具有对称性,在逼近实际信号先验分布时会受到限制;而期望最大化高斯混合近似信息传递（EM-GM-AMP）算法中的GM模型是BG模型的高阶形式,复杂度较高。为了解决以上问题,提出贝努利不对称高斯模型（BAG）,进而推导得到期望最大化贝努利不对称高斯近似信息传递（EM-BAG-AMP）算法。该算法的主要思路是假设输入信号服从BAG模型,然后使用广义近似信息传递（GAMP）重构信号并在算法迭代中同时更新模型参数。实验证明,在处理不同图像数据时,EM-BAG-AMP和EM-BG-AMP相比,时间增加了1.2%,峰值信噪比（PSNR）值提升了0.1~0.5 d B,尤其在处理纹理较少以及色差变化明显的图像时峰值信噪比（PSNR）值提升了0.4~0.5 d B。EM-BAG-AMP是对EM-BG-AMP算法的扩展和延伸,更适合实际信号的处理。展开更多
文摘A fuzzy modeling method for complex systems is studied. The notation of general stochastic neural network (GSNN) is presented and a new modeling method is given based on the combination of the modified Takagi and Sugeno's (MTS) fuzzy model and one-order GSNN. Using expectation-maximization(EM) algorithm, parameter estimation and model selection procedures are given. It avoids the shortcomings brought by other methods such as BP algorithm, when the number of parameters is large, BP algorithm is still difficult to apply directly without fine tuning and subjective tinkering. Finally, the simulated example demonstrates the effectiveness.
文摘期望最大化贝努利高斯（BG）近似信息传递（EM-BG-AMP）算法中的BG模型因为具有对称性,在逼近实际信号先验分布时会受到限制;而期望最大化高斯混合近似信息传递（EM-GM-AMP）算法中的GM模型是BG模型的高阶形式,复杂度较高。为了解决以上问题,提出贝努利不对称高斯模型（BAG）,进而推导得到期望最大化贝努利不对称高斯近似信息传递（EM-BAG-AMP）算法。该算法的主要思路是假设输入信号服从BAG模型,然后使用广义近似信息传递（GAMP）重构信号并在算法迭代中同时更新模型参数。实验证明,在处理不同图像数据时,EM-BAG-AMP和EM-BG-AMP相比,时间增加了1.2%,峰值信噪比（PSNR）值提升了0.1~0.5 d B,尤其在处理纹理较少以及色差变化明显的图像时峰值信噪比（PSNR）值提升了0.4~0.5 d B。EM-BAG-AMP是对EM-BG-AMP算法的扩展和延伸,更适合实际信号的处理。
