On the Non-Asymptotic Concentration of Heteroskedastic Wishart-type Matrix

preprint2022arXivOpen access

Abstract

This paper focuses on the non-asymptotic concentration of the heteroskedastic Wishart-type matrices. Suppose ZZ is a p1p_1-by-p2p_2 random matrix and Zij∼N(0,σij2)Z_{ij} \sim N(0,σ_{ij}^2) independently, we prove the expected spectral norm of Wishart matrix deviations (i.e., E∥ZZ⊤−EZZ⊤∥\mathbb{E} \left\|ZZ^\top - \mathbb{E} ZZ^\top\right\|) is upper bounded by \begin{equation*} \begin{split} (1+ε)\left\{2σ_Cσ_R + σ_C^2 + Cσ_Rσ_*\sqrt{\log(p_1 \wedge p_2)} + Cσ_*^2\log(p_1 \wedge p_2)\right\}, \end{split} \end{equation*} where σC2:=max⁡j∑i=1p1σij2σ_C^2 := \max_j \sum_{i=1}^{p_1}σ_{ij}^2, σR2:=max⁡i∑j=1p2σij2σ_R^2 := \max_i \sum_{j=1}^{p_2}σ_{ij}^2 and σ∗2:=max⁡i,jσij2σ_*^2 := \max_{i,j}σ_{ij}^2. A minimax lower bound is developed that matches this upper bound. Then, we derive the concentration inequalities, moments, and tail bounds for the heteroskedastic Wishart-type matrix under more general distributions, such as sub-Gaussian and heavy-tailed distributions. Next, we consider the cases where ZZ has homoskedastic columns or rows (i.e., σij≈σiσ_{ij} \approx σ_i or σij≈σjσ_{ij} \approx σ_j) and derive the rate-optimal Wishart-type concentration bounds. Finally, we apply the developed tools to identify the sharp signal-to-noise ratio threshold for consistent clustering in the heteroskedastic clustering problem.

Explore this paper’s authors, topics and related work.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Reviews 0

Write a reviewWrite

No reviews yet.

Discussion 0

Add a commentWrite

No comments yet.