Deformed semicircle law and concentration of nonlinear random matrices for ultra-wide neural networks (Q6590448)

scientific article; zbMATH DE number 7899436

Language	Label	Description	Also known as
English	Deformed semicircle law and concentration of nonlinear random matrices for ultra-wide neural networks	scientific article; zbMATH DE number 7899436

Statements

instance of

scholarly article

0 references

title

Deformed semicircle law and concentration of nonlinear random matrices for ultra-wide neural networks (English)

0 references

0 references

0 references

The Annals of Applied Probability

0 references

publication date

21 August 2024

0 references

zbMATH Keywords

neural networks

0 references

neural tangent kernel

0 references

random feature regression

0 references

random matrix theory

0 references

MaRDI profile type

Publication

0 references

cites work

A note on the Hanson-Wright inequality for random vectors with dependencies

0 references

PARTIAL TRANSPOSITION OF RANDOM STATES AND NON-CENTERED SEMICIRCULAR DISTRIBUTIONS

0 references

Alice and Bob Meet Banach

0 references

On the Equivalence between Kernel Quadrature Rules and Random Feature Expansions

0 references

Spectral analysis of large dimensional random matrices

0 references

Convergence to the semicircle law

0 references

The limiting spectral distribution of the product of the Wigner matrix and a nonnegative definite matrix

0 references

Strong convergence of ESD for the generalized sample covariance matrices when \(p/n \rightarrow 0\)

0 references

An Inverse Matrix Adjustment Arising in Discriminant Analysis

0 references

Deep learning: a statistical viewpoint

0 references

Eigenvalue distribution of some nonlinear models of random matrices

0 references

Q4907706

0 references

CLT for linear spectral statistics of normalized sample covariance matrices with the dimension much larger than the sample size

0 references

Convergence of the largest eigenvalue of normalized sample covariance matrices when \(p\) and \(n\) both tend to infinity with their ratio converging to zero

0 references

Asymptotic freeness of layerwise Jacobians caused by invariance of multilayer perceptron: the Haar orthogonal case

0 references

The PPT square conjecture holds generically for some classes of independent states

0 references

Spiked singular values and vectors under extreme aspect ratios

0 references

Large-dimensional random matrix theory and its applications in deep learning and wireless communications

0 references

A Bound on Tail Probabilities for Quadratic Forms in Independent Random Variables

0 references

Universality Laws for High-Dimensional Learning With Random Features

0 references

Q3580453

0 references

Q3996929

0 references

Testing the sphericity of a covariance matrix when the dimension is much larger than the sample size

0 references

Just interpolate: kernel ``ridgeless'' regression can generalize

0 references

A random matrix approach to neural networks

0 references

Learning curves of generic features maps for realistic datasets with a teacher-student model*

0 references

Generalization error of random feature and kernel methods: hypercontractivity and kernel matrix concentration

0 references

The Generalization Error of Random Features Regression: Precise Asymptotics and the Double Descent Curve

0 references

The interpolation phase transition in neural networks: memorization and generalization under lazy training

0 references

Lectures on the Combinatorics of Free Probability

0 references

A note on the Pennington-Worah distribution

0 references

Asymptotic normality for eigenvalue statistics of a general sample covariance matrix when \(p/n \to \infty\) and applications

0 references

Hanson-Wright inequality and sub-Gaussian concentration

0 references

The smallest eigenvalue of a large dimensional Wishart matrix

0 references

User-friendly tail bounds for sums of random matrices

0 references

High-Dimensional Probability

0 references

Q3811203

0 references

Limiting spectral distribution of renormalized separable sample covariance matrices when \(p/n\to 0\)

0 references

Limiting spectral distribution of normalized sample covariance matrices with \(p/n\to 0\)

0 references

Identifiers

DOI

10.1214/23-aap2010

0 references

Mathematics Subject Classification ID

0 references

0 references

0 references

0 references

Sitelinks

Mathematics(1 entry)

mardi Publication:6590448