Designing stable neural networks using convex analysis and ODEs
DOI10.1016/j.physd.2024.134159zbMATH Open1541.65055MaRDI QIDQ6554924
Matthias J. Ehrhardt, Brynjulf Owren, Carola-Bibiane Schönlieb, Ferdia Sherry, Elena Celledoni, Davide Murari
Publication date: 13 June 2024
Published in: Physica D (Search for Journal in Brave)
Artificial neural networks and deep learning (68T07) Monotone operators and generalizations (47H05) Nonlinear ordinary differential equations and systems (34A34) Numerical methods for initial value problems involving ordinary differential equations (65L05) Numerical quadrature and cubature formulas (65D32) Applications of operator theory in optimization, convex analysis, mathematical programming, economics (47N10)
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Nonlinear total variation based noise removal algorithms
- Parseval proximal neural networks
- Eigenvalue computation in the 20th century
- Deep neural networks motivated by partial differential equations
- Deep neural network structures solving variational inequalities
- Convolutional proximal neural networks and plug-and-play algorithms
- Reducibility and contractivity of Runge-Kutta methods revisited
- Solving Ordinary Differential Equations I
- A Nonexistence Theorem for Explicit A-Stable Methods
- Stability Criteria for Implicit Runge–Kutta Methods
- Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
- Contractivity of Runge--Kutta Methods for Convex Gradient Systems
- Learning Maximally Monotone Operators for Image Recovery
- Convex analysis and monotone operator theory in Hilbert spaces
- Deep neural networks on diffeomorphism groups for optimal shape reparametrization
- Dynamical Systems–Based Neural Networks
Related Items (1)
This page was built for publication: Designing stable neural networks using convex analysis and ODEs
