论文信息 - Generalized Independent Component Analysis as Density Estimation

Generalized Independent Component Analysis as Density Estimation

We propose a new generalized ICA frameworkin the form of a multi-layer perceptron as a density estimator. We adopt an optimization strategy based on two criteria: a minimum reconstruction error and a minimum distance from a uniform distribution. Some simulation results are also reported to validate the proposed algorithm.

Francesco Palmieri | Alessandra Budillon

[1] M. Kramer. Nonlinear principal component analysis using autoassociative neural networks , 1991 .

[2] F. Palmieri,et al. Multi-Class Independent Component Analysis (MUCICA) for Rank-Deficient Distributions , 2000 .

[3] Juha Karhunen,et al. Representation and separation of signals using nonlinear PCA type learning , 1994, Neural Networks.

[4] Christian Jutten,et al. Source separation in post-nonlinear mixtures , 1999, IEEE Trans. Signal Process..