中文说明：
这是一个贝叶斯独立分量分析(ICA)算法的线性瞬时混合高斯噪声模型和添加剂。解决问题的是ML-II推论,即资源的整合在发现源后和噪声协方差矩阵和混合了最大化的边际似然。充分统计量的估计平均场或变分理论和线性响应修正或通过自适应平均场理论水龙头。平均场方程,解决了信仰传播法的或连续的迭代。
English Description：
This is a Bayesian independent component analysis (ICA) algorithm of linear instantaneous Gaussian mixture noise model and additive. The solution to the problem is ml-ii inference, that is, the integration of resources and the noise covariance matrix and the maximum marginal likelihood are mixed after the source is found. Sufficient statistics are estimated by mean field or variational theory and linear response correction or by adaptive mean field theory. The mean field equation solves the belief propagation method or continuous iteration.