Fourier PCA and Robust Tensor Decomposition
Navin Goyal, Santosh Vempala, Ying Xiao
(Submitted on 25 Jun 2013 - revised 27 Jun 2014 (this version, v5))
Fourier PCA is Principal Component Analysis of a matrix obtained from higher order derivatives of the logarithm of the Fourier transform of a distribution.We make this method algorithmic by developing a tensor decomposition method for a pair of tensors sharing the same vectors in rank-1 decompositions. Our main application is the first provably polynomial-time algorithm for underdetermined ICA, i.e., learning an n×mmatrix A from observations y=Ax where x is drawn from an unknown product distribution with arbitrary non-Gaussian components. The number of component distributions m can be arbitrarily higher than the dimension n and the columns of A only need to satisfy a natural and efficiently verifiable nondegeneracy condition. As a second application, we give an alternative algorithm for learning mixtures of spherical Gaussians with linearly independent means. These results also hold in the presence of Gaussian noise.