Estimating Rank-One Spikes from Heavy-Tailed Noise via Self-Avoiding Walks
NIPS 2020, 2020.
First we show that for many long tail distributions, Principal component analysis algorithm can be saved by such truncation.
We study symmetric spiked matrix models with respect to a general class of noise distributions. Given a rank-1 deformation of a random noise matrix, whose entries are independently distributed with zero mean and unit variance, the goal is to estimate the rank-1 part. For the case of Gaussian noise, the top eigenvector of the given matri...More
PPT (Upload PPT)