Tail Bounds on the Spectral Norm of Sub-Exponential Random Matrices

DOI10.1142/S2010326323500132MaRDI QIDQ6420589zbMATH OpenOpenAlexFDO

Authors Guozheng Dai, Zhonggen Su, Hanchao Wang

Publication date 14 December 2022

Full work available at URL https://doi.org/10.1142/s2010326323500132

Mathematics Subject Classification ID

Random matrices (algebraic aspects) (15B52) Probabilistic methods in Banach space theory (46B09) Random matrices (probabilistic aspects) (60B20)

Abstract: Let $X$ be an $nimesn$ symmetric random matrix with independent but non-identically distributed entries. The deviation inequalities of the spectral norm of $X$ with Gaussian entries have been obtained by using the standard concentration of measure results in Gauss space which is unfortunately not suitable for the sub-Exponential case. This paper establishes an upper tail bound of $VertXVert$ with sub-Exponential entries. Our general method relies upon a crucial ingredient of a novel chaining argument that essentially depends on the distribution of coordinates of a point on the unit sphere. What makes this approach work is the particular structure of the sets used for the chaining.

Recommendations

• Sharp nonasymptotic bounds on the norm of random matrices with independent entries
• On the lower bound of the spectral norm of symmetric random matrices with independent entries
• Spectral norm of random matrices
• User-friendly tail bounds for sums of random matrices
• Spectral norm of products of random and deterministic matrices

Cited in

(1)

• Differentially private low-dimensional synthetic data from high-dimensional datasets

This page was built for publication: Tail Bounds on the Spectral Norm of Sub-Exponential Random Matrices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6420589)