论文部分内容阅读
自从Ｌｉｎｓｋｅｒ提出生物系统对信息的处理可能遵循信息传输最大化准则（Ｉｎｆｏｍａｘ）以来，Ｉｎｆｏｍａｘ算法已被成功地应用于很多领域。作者在“非线性Ｉｎｆｏｒｍａｘ自组织算法的盲源分离机理”一文中研究了输入－输出为非线性映射、无输入噪声时Ｉｎｆｏｍａｘ算法的性能，本文在以前工作的基础上，详细讨论了在输入－输出为线性映射、输入信号和噪声为高斯分布条件下Ｉｎｆｏｍａｘ算法的性能，并与传统的ＰＣＡ算法作了比较分析。
The Infomax algorithm has been successfully applied in many fields since Linsker proposed that biological systems should process information in accordance with the maximization of information transmission (Infomax). The author studied the performance of Infomax algorithm with nonlinear input-output mapping and no input noise based on the “blind source separation mechanism of nonlinear Informax self-organizing algorithm”. Based on the previous work, The output is the performance of Infomax algorithm with linear mapping, input signal and noise Gaussian distribution, and compared with the traditional PCA algorithm.