Deep limits of residual neural networks

DOI10.1007/s40687-022-00370-yOpenAlexW2898318847MaRDI QIDQ2679108

Yves van Gennip, Matthew Thorpe

Publication date: 19 January 2023

Published in: Research in the Mathematical Sciences (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1810.11741

zbMATH Keywords

regularityordinary differential equationsgamma-convergencevariational convergencedeep neural networksdeep layer limits

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Neural networks for/in biological studies, artificial life and related topics (92B20) Methods involving semicontinuity and convergence; relaxation (49J45) Existence theories for optimal control problems involving ordinary differential equations (49J15) Applications of difference equations (39A60)

Related Items (4)

Rates of convergence for Laplacian semi-supervised learning with low labeling rates ⋮ Dynamical Systems–Based Neural Networks ⋮ Forward stability of ResNet and its variants ⋮ Deep learning as optimal control problems: models and numerical methods

Uses Software

• ImageNet
• FFJORD
• AlexNet

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Continuum limit of total variation on point clouds
• Local minimization, variational evolution and \(\Gamma\)-convergence
• A variational approach to the consistency of spectral clustering
• An introduction to \(\Gamma\)-convergence
• Multilayer feedforward networks are universal approximators
• \(\Gamma\)-convergence of graph Ginzburg-Landau functionals
• Forward stability of ResNet and its variants
• Deep neural networks motivated by partial differential equations
• Deep relaxation: partial differential equations for optimizing deep neural networks
• A mean-field optimal control formulation of deep learning
• Blended coarse gradient descent for full quantization of deep neural networks
• A proposal on machine learning via dynamical systems
• Discrete Mathematics of Neural Networks
• Consistency of Cheeger and Ratio Graph Cuts
• Analysis of recursive stochastic algorithms
• A Mathematical Theory of Deep Convolutional Neural Networks for Feature Extraction
• Stable architectures for deep neural networks
• Partial differential equation regularization for supervised machine learning
• Structure-preserving deep learning
• Neural networks and complexity theory
• Ensemble Kalman inversion: a derivative-free technique for machine learning tasks
• Analysis of $p$-Laplacian Regularization in Semisupervised Learning
• Deep Learning: An Introduction for Applied Mathematicians
• On a Stochastic Approximation Method
• Approximation by superpositions of a sigmoidal function
• PDE-based group equivariant convolutional neural networks

This page was built for publication: Deep limits of residual neural networks