Part of Advances in Neural Information Processing Systems 10 (NIPS 1997)
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 decompo(cid:173) sition. Using only the local geometric structure of a class repre(cid:173) sentative, we derive an analytic solution for the factorization. We demonstrate the factorization solution with numerical experiments and present a preliminary tie to decorrelation.