Independent component analysis, a new concept?
Signal Processing - Special issue on higher order statistics
Independent component analysis for identification of artifacts in magnetoencephalographic recordings
NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
New approximations of differential entropy for independent component analysis and projection pursuit
NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
Independent components of magnetoencephalography: localization
Neural Computation
Topographic Independent Component Analysis
Neural Computation
Complexity Pursuit: Separating Interesting Components from Time Series
Neural Computation
A blind source separation technique using second-order statistics
IEEE Transactions on Signal Processing
Fast and robust fixed-point algorithms for independent component analysis
IEEE Transactions on Neural Networks
Comparison of BSS methods for the detection of α-activity components in EEG
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Blind source separation of cardiac murmurs from heart recordings
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Brains and phantoms: an ICA study of fMRI
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Hi-index | 0.00 |
The present paper is written as a word of caution, with users of independent component analysis (ICA) in mind, to overlearning phenomena that are often observed.We consider two types of overlearning, typical to high-order statistics based ICA. These algorithms can be seen to maximise the negentropy of the source estimates. The first kind of overlearning results in the generation of spike-like signals, if there are not enough samples in the data or there is a considerable amount of noise present. It is argued that, if the data has power spectrum characterised by 1/f curve, we face a more severe problem, which cannot be solved inside the strict ICA model. This overlearning is better characterised by bumps instead of spikes. Both overlearning types are demonstrated in the case of artificial signals as well as magnetoencephalograms (MEG). Several methods are suggested to circumvent both types, either by making the estimation of the ICA model more robust or by including further modelling of the data.