Factorizing Multivariate Function Classes
–Neural Information Processing Systems
The mathematical framework for factorizing equivalence classes of multivariate functions is formulated in this paper. Independent component analysis is shown to be a special case of this decomposition. Using only the local geometric structure of a class representative, we derive an analytic solution for the factorization. We demonstrate the factorization solution with numerical experiments and present a preliminary tie to decorrelation.
Neural Information Processing Systems
Dec-31-1998
- Country:
- North America > United States > Massachusetts > Middlesex County > Cambridge (0.15)
- Technology: