Overall error analysis for the training of deep neural networks via stochastic gradient descent with random initialisation (Q6107984)

scientific article; zbMATH DE number 7704193

Language	Label	Description	Also known as
English	Overall error analysis for the training of deep neural networks via stochastic gradient descent with random initialisation	scientific article; zbMATH DE number 7704193

Statements

instance of

scholarly article

0 references

title

Overall error analysis for the training of deep neural networks via stochastic gradient descent with random initialisation (English)

0 references

0 references

0 references

Applied Mathematics and Computation

0 references

publication date

29 June 2023

0 references

full work available at URL

https://arxiv.org/abs/2003.01291

0 references

zbMATH Keywords

deep learning

0 references

deep neural networks

0 references

empirical risk minimisation

0 references

full error analysis

0 references

approximation

0 references

generalisation

0 references

optimisation

0 references

strong convergence

0 references

stochastic gradient descent

0 references

random initialisation

0 references

MaRDI profile type

MaRDI publication profile

0 references

cites work

Full error analysis for the training of deep neural networks

0 references

Approximation by superpositions of a sigmoidal function

0 references

Multilayer feedforward networks are universal approximators

0 references

Universal approximation bounds for superpositions of a sigmoidal function

0 references

Approximation and estimation bounds for artificial neural networks

0 references

Neural Networks for Localized Approximation

0 references

Q4310147

0 references

Q4944751

0 references

On the mathematical foundations of learning

0 references

A distribution-free theory of nonparametric regression

0 references

Local Rademacher complexities

0 references

Understanding Machine Learning

0 references

Analysis of the Generalization Error: Empirical Risk Minimization over Deep Artificial Neural Networks Overcomes the Curse of Dimensionality in the Numerical Approximation of Black--Scholes Partial Differential Equations

0 references

A priori estimates of the population risk for two-layer neural networks

0 references

A comparative analysis of optimization and generalization properties of two-layer neural network and random feature models under gradient descent dynamics

0 references

General multilevel adaptations for stochastic approximation algorithms of Robbins-Monro and Polyak-Ruppert type

0 references

Q4969246

0 references

Strong error analysis for stochastic gradient descent optimization algorithms

0 references

Lower error bounds for the stochastic gradient descent optimization algorithm: sharp convergence rates for slowly and fast decaying learning rates

0 references

Gradient descent optimizes over-parameterized deep ReLU networks

0 references

Q3241581

0 references

Tractability of multivariate problems. Volume I: Linear information

0 references

Tractability of multivariate problems. Volume II: Standard information for functionals.

0 references

A Stochastic Approximation Method

0 references

Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations

0 references

Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations

0 references

Q5270493

0 references

Space-time error estimates for deep neural network approximations for differential equations

0 references

DGM: a deep learning algorithm for solving partial differential equations

0 references

Deep neural network approximations for solutions of PDEs based on Monte Carlo algorithms

0 references

On the approximation by single hidden layer feedforward neural networks with fixed weights

0 references

Breaking the Curse of Dimensionality with Convex Neural Networks

0 references

Optimal Approximation with Sparsely Connected Deep Neural Networks

0 references

Error bounds for approximation with neural networks

0 references

Q4365433

0 references

Exponential convergence of the deep neural network approximation for analytic functions

0 references

DNN expression rate analysis of high-dimensional PDEs: application to option pricing

0 references

A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations

0 references

Proof that deep artificial neural networks overcome the curse of dimensionality in the numerical approximation of Kolmogorov partial differential equations with constant diffusion and nonlinear drift coefficients

0 references

Degree of approximation by neural and translation networks with a single hidden layer

0 references

Deep vs. shallow networks: An approximation theory perspective

0 references

New Error Bounds for Deep ReLU Networks Using Sparse Grids

0 references

Approximation of functions and their derivatives: A neural network implementation with applications

0 references

Optimal approximation of piecewise smooth functions using deep ReLU neural networks

0 references

Q4938227

0 references

Deep learning in high dimension: Neural network expression rates for generalized polynomial chaos expansions in UQ

0 references

Provable approximation properties for deep neural networks

0 references

Nonlinear approximation via compositions

0 references

Error bounds for approximations with deep ReLU networks

0 references

Convergence in Hölder norms with applications to Monte Carlo methods in infinite dimensions

0 references

Uniform error estimates for artificial neural network approximations for heat equations

0 references

Some Elementary Inequalities Relating to the Gamma and Incomplete Gamma Function

0 references

Bounds for the ratio of two gamma functions

0 references

Q5386174

0 references

Identifiers

arXiv ID

2003.01291

0 references

DOI

10.1016/j.amc.2023.127907

0 references

Mathematics Subject Classification ID

0 references

0 references

0 references

0 references

0 references

Sitelinks

Mathematics(1 entry)

mardi Publication:6107984