Subspace power method for symmetric tensor decomposition and generalized PCA

12/09/2019
by   Joe Kileel, et al.
0

We introduce the Subspace Power Method (SPM) for decomposing symmetric even-order tensors. This algorithm relies on a power method iteration applied to a modified tensor, constructed from a matrix flattening. For tensors of rank up to roughly the square root of the number of tensor entries, we obtain provable guarantees for most steps of SPM, by drawing on results from algebraic geometry and dynamical systems. Numerical simulations indicate that SPM significantly outperforms state-of-the-art algorithms in terms of speed, while performing robustly for low-rank tensors subjected to additive noise. We also extend SPM to compute a certain generalized tensor decomposition, enabling generalized principal component analysis at extremely low SNR.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 32

11/12/2021

Robust Eigenvectors of Symmetric Tensors

The tensor power method generalizes the matrix power method to higher or...
03/13/2018

Principal Component Analysis with Tensor Train Subspace

Tensor train is a hierarchical tensor network structure that helps allev...
10/29/2021

Landscape analysis of an improved power method for tensor decomposition

In this work, we consider the optimization formulation for symmetric ten...
05/02/2018

Computing tensor Z-eigenvectors with dynamical systems

We present a new framework for computing Z-eigenvectors of general tenso...
11/21/2017

Kullback-Leibler Principal Component for Tensors is not NP-hard

We study the problem of nonnegative rank-one approximation of a nonnegat...
12/29/2020

Inference for Low-rank Tensors – No Need to Debias

In this paper, we consider the statistical inference for several low-ran...
05/10/2020

Robust Tensor Decomposition for Image Representation Based on Generalized Correntropy

Traditional tensor decomposition methods, e.g., two dimensional principa...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.