Independent Subspace Analysis Is Unique, Given Irreducibility
- 2k Downloads
Independent Subspace Analysis (ISA) is a generalization of ICA. It tries to find a basis in which a given random vector can be decomposed into groups of mutually independent random vectors. Since the first introduction of ISA, various algorithms to solve this problem have been introduced, however a general proof of the uniqueness of ISA decompositions remained an open question. In this contribution we address this question and sketch a proof for the separability of ISA. The key condition for separability is to require the subspaces to be not further decomposable (irreducible). Based on a decomposition into irreducible components, we formulate a general model for ISA without restrictions on the group sizes. The validity of the uniqueness result is illustrated on a toy example. Moreover, an extension of ISA to subspace extraction is introduced and its indeterminacies are discussed.
KeywordsRandom Vector Independent Component Irreducible Component Independent Component Analysis Invertible Matrix
Unable to display preview. Download preview PDF.
- 1.Blanchard, G., Kawanabe, M., Sugiyama, M., Spokoiny, V., Müller, K.R.: In search of non-gaussian components of a high-dimensional distribution. Journal of Machine Learning Research 7, 247–282 (2006)Google Scholar
- 2.Cardoso, J.F.: Multidimensional independent component analysis. In: Proc. of ICASSP 1998, Seattle (1998)Google Scholar
- 5.Theis, F.J.: Towards a general independent subspace analysis. In: Proc. NIPS 2006 (2007)Google Scholar