Loading…

The Landscape of the Spiked Tensor Model

We consider the problem of estimating a large rank‐one tensor u⊗k ∈ (ℝn)⊗k, k ≥ 3, in Gaussian noise. Earlier work characterized a critical signal‐to‐noise ratio λ  Bayes = O(1) above which an ideal estimator achieves strictly positive correlation with the unknown vector of interest. Remarkably, no...

Full description

Saved in:
Bibliographic Details
Published in:Communications on pure and applied mathematics 2019-11, Vol.72 (11), p.2282-2330
Main Authors: Arous, Gérard Ben, Mei, Song, Montanari, Andrea, Nica, Mihai
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We consider the problem of estimating a large rank‐one tensor u⊗k ∈ (ℝn)⊗k, k ≥ 3, in Gaussian noise. Earlier work characterized a critical signal‐to‐noise ratio λ  Bayes = O(1) above which an ideal estimator achieves strictly positive correlation with the unknown vector of interest. Remarkably, no polynomial‐time algorithm is known that achieved this goal unless λ ≥ Cn(k − 2)/4, and even powerful semidefinite programming relaxations appear to fail for 1 ≪ λ ≪ n(k − 2)/4. In order to elucidate this behavior, we consider the maximum likelihood estimator, which requires maximizing a degree‐k homogeneous polynomial over the unit sphere in n dimensions. We compute the expected number of critical points and local maxima of this objective function and show that it is exponential in the dimensions n, and give exact formulas for the exponential growth rate. We show that (for λ larger than a constant) critical points are either very close to the unknown vector u or are confined in a band of width Θ(λ−1/(k − 1)) around the maximum circle that is orthogonal to u. For local maxima, this band shrinks to be of size Θ(λ−1/(k − 2)). These “uninformative” local maxima are likely to cause the failure of optimization algorithms. © 2019 Wiley Periodicals, Inc.
ISSN:0010-3640
1097-0312
DOI:10.1002/cpa.21861