Interpolation and approximation via momentum ResNets and neural ODEs

DOI10.1016/j.sysconle.2022.105182zbMath1485.93243arXiv2110.08761OpenAlexW3207108121MaRDI QIDQ2124500

Elisa Affili, Domènec Ruiz-Balet, Enrique Zuazua

Publication date: 11 April 2022

Published in: Systems \& Control Letters (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2110.08761

zbMATH Keywords

universal approximationdeep learningsimultaneous controlmomentum ResNetsneural differential equationssimultaneous tracking controllability

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Observability (93B07) Control/observation systems governed by ordinary differential equations (93C15)

Related Items (5)

Turnpike in optimal control of PDEs, ResNets, and beyond ⋮ Sparsity in long-time control of neural ODEs ⋮ Neural ODE Control for Classification, Approximation, and Transport ⋮ Control of neural transport for normalising flows ⋮ Control of partial differential equations via physics-informed neural networks

Uses Software

• torchdiffeq
• BiT

Cites Work

• Unnamed Item
• From averaged to simultaneous controllability
• Null controllability of a system of viscoelasticity with a moving control
• Multilayer feedforward networks are universal approximators
• Approximation rates for neural networks with encodable weights in smoothness spaces
• Averaged dynamics and control for heat equations with random diffusion
• Control under constraints for multi-dimensional reaction-diffusion monostable and bistable equations
• A proposal on machine learning via dynamical systems
• Null controllability for wave equations with memory
• Exact Controllability, Stabilization and Perturbations for Distributed Systems
• Stable architectures for deep neural networks
• Turnpike in Lipschitz—nonlinear optimal control
• Deep Neural Networks, Generic Universal Interpolation, and Controlled ODEs
• Controllability of Evolution Equations with Memory
• Approximation by superpositions of a sigmoidal function
• Controllability of the heat equation with memory

This page was built for publication: Interpolation and approximation via momentum ResNets and neural ODEs