A Michigan-like immune-inspired framework for performing independent component analysis over Galois fields of prime order

An Introduction to Information Theoretic Learning, Part II: Applications

“…Another related proposal was subsequently developed [58], where a more robust immune-inspired algorithm was applied in association to a Michigan-like approach [59] to model the population individuals. The algorithm criterion was to minimize the entropy of each extracted source, considering that the intrinsic diversity operators of the algorithm may allow that distinct independent signals are obtained, in the end.…”

Blind Source Separation: Fundamentals and Perspectives on Galois Fields and Sparse Signals

“…A different perspective lies on considering the same criterion with lower-cost metaheuristics that are appealing for combinatorial problems, e.g. Artificial Immune Systems (AIS) [24]. In this case, the algorithm optimizes (13), but at the end of the procedure, the N best candidate-solutions which are linearly independent represent the extraction vectors that, finally, compose the separating matrix.…”

A study on independent component analysis over galois fields

“…Hence, we consider still valid the hypothetical explanation presented by SILVA et al, (2014a), where the authors referred to WATERHOUSE (1987), who has shown that, based on Equation 2.5, the probability that a matrix ( ) in ( ) has null determinant is In order to analyze the sensitivity to the non-uniform restriction (recall Section 2.2), The results reinforce that ICA over GF only works for non-uniform distributions, as already mentioned in Section 2.2, such that, the higher is the threshold, the more "nonuniform" the sources are generated.…”

“…If any G ( ) could be found such that y = Gs be composed of independent components again, then G = PD, for some permutation matrix P and some diagonal matrix SILVA et al, 2014a).…”

“…In order to find an hypothesis for this behavior we refer to the results obtained by a Michigan-like 1 implementation of cob-aiNet[C] for ICA over GF in (SILVA et al, 2014a),…”