Near-Optimal Tensor PCA via Normalized Stochastic Gradient Ascent with Overparameterization

Shihong Ding, Yihong Gu, Yuanshi Liu +1

We study the Order-$k$ ($k\ge 4$) spiked tensor model for the tensor principal component analysis (PCA) problem: given $N$ i.i.d. observations of a $k$-th order tensor generated…