Non-convergence of stochastic gradient descent in the training of deep neural networks

DOI10.1016/j.jco.2020.101540zbMath1494.65044arXiv2006.07075OpenAlexW3107512664MaRDI QIDQ2034567

Patrick Cheridito, Florian Rossmannek, Arnulf Jentzen

Publication date: 22 June 2021

Published in: Journal of Complexity (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2006.07075

zbMATH Keywords

machine learningempirical risk minimizationnon-convergencedeep neural networksstochastic gradient descend

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Numerical optimization and variational techniques (65K10)

Related Items (6)

Stationary Density Estimation of Itô Diffusions Using Deep Learning ⋮ Solving high-dimensional Hamilton-Jacobi-Bellman PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space ⋮ A proof of convergence for gradient descent in the training of artificial neural networks for constant target functions ⋮ A proof of convergence for stochastic gradient descent in the training of artificial neural networks with ReLU activation for constant target functions ⋮ Deep multimodal autoencoder for crack criticality assessment ⋮ Constructive deep ReLU neural network approximation

Cites Work

• Gradient descent optimizes over-parameterized deep ReLU networks
• A comparative analysis of optimization and generalization properties of two-layer neural network and random feature models under gradient descent dynamics
• Theoretical Insights Into the Optimization Landscape of Over-Parameterized Shallow Neural Networks
• Optimization Methods for Large-Scale Machine Learning

This page was built for publication: Non-convergence of stochastic gradient descent in the training of deep neural networks