Forward stability of ResNet and its variants

DOI10.1007/s10851-019-00922-yzbMath1434.68528arXiv1811.09885OpenAlexW2986677756WikidataQ126842178 ScholiaQ126842178MaRDI QIDQ1988347

Linan Zhang, Hayden Schaeffer

Publication date: 23 April 2020

Published in: Journal of Mathematical Imaging and Vision (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1811.09885

zbMATH Keywords

stabilitydifferential inclusionsoptimal control problemsdeep feedforward neural networksresidual neural networks

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Sensitivity, stability, well-posedness (49K40) Ordinary differential inclusions (34A60)

Related Items (8)

Translating numerical concepts for PDEs into neural architectures ⋮ Deep learning for inverse problems. Abstracts from the workshop held March 7--13, 2021 (hybrid meeting) ⋮ Designing rotationally invariant neural networks from PDEs and variational methods ⋮ Deep limits of residual neural networks ⋮ A physics-constrained deep residual network for solving the sine-Gordon equation ⋮ Connections between numerical algorithms for PDEs and neural networks ⋮ Quantized convolutional neural networks through the lens of partial differential equations ⋮ Structure-preserving deep learning

Uses Software

• ImageNet
• FASTA
• AlexNet
• Entropy-SGD

Cites Work

• Unnamed Item
• Unnamed Item
• A penalty method for some nonlinear variational obstacle problems
• Global stability of almost periodic solutions to monotone sweeping processes and their response to non-monotone perturbations
• Deep relaxation: partial differential equations for optimizing deep neural networks
• A mean-field optimal control formulation of deep learning
• A proposal on machine learning via dynamical systems
• Relaxation of an optimal control problem involving a perturbed sweeping process
• Deep limits of residual neural networks
• Learning Deep Architectures for AI
• Splitting Algorithms for the Sum of Two Nonlinear Operators
• Stable architectures for deep neural networks
• Optimization Methods for Large-Scale Machine Learning
• Nonsmooth sequential analysis in Asplund spaces
• Prox-regular functions in variational analysis
• An $L^1$ Penalty Method for General Obstacle Problems
• Entropy-SGD: biasing gradient descent into wide valleys

This page was built for publication: Forward stability of ResNet and its variants