SPLICE: Fully Tractable Hierarchical Extension of ICA with Pooling
Jun-ichiro Hirayama, Aapo Hyvärinen, Motoaki Kawanabe
Abstract
We present a novel probabilistic framework for a hierarchical extension of independent component analysis (ICA), with a particular motivation in neuroscientific data analysis and modeling. The framework incorporates a general subspace pooling with linear ICA-like layers stacked recursively. Unlike related previous models, our generative model is fully tractable: both the likelihood and the posterior estimates of latent variables can readily be computed with analytically simple formulae. The model is particularly simple in the case of complex-valued data since the pooling can be reduced to taking the modulus of complex numbers. Experiments on electroencephalography (EEG) and natural images demonstrate the validity of the method.
BibTeX
@InProceedings{pmlr-v70-hirayama17a,
title = {{SPLICE}: Fully Tractable Hierarchical Extension of {ICA} with Pooling},
author = {Jun-ichiro Hirayama and Aapo Hyv{\"a}rinen and Motoaki Kawanabe},
booktitle = {Proceedings of the 34th International Conference on Machine Learning},
pages = {1491--1500},
year = {2017},
editor = {Precup, Doina and Teh, Yee Whye},
volume = {70},
series = {Proceedings of Machine Learning Research},
month = {06--11 Aug},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v70/hirayama17a/hirayama17a.pdf},
url = {https://proceedings.mlr.press/v70/hirayama17a.html},
abstract = {We present a novel probabilistic framework for a hierarchical extension of independent component analysis (ICA), with a particular motivation in neuroscientific data analysis and modeling. The framework incorporates a general subspace pooling with linear ICA-like layers stacked recursively. Unlike related previous models, our generative model is fully tractable: both the likelihood and the posterior estimates of latent variables can readily be computed with analytically simple formulae. The model is particularly simple in the case of complex-valued data since the pooling can be reduced to taking the modulus of complex numbers. Experiments on electroencephalography (EEG) and natural images demonstrate the validity of the method.}
}