Continuous-time system identification with neural networks: model structures and fitting criteria

DOI10.1016/j.ejcon.2021.01.008zbMath1466.93032arXiv2006.02915OpenAlexW3033855706MaRDI QIDQ2034176

Marco Forgione, Dario Piga

Publication date: 21 June 2021

Published in: European Journal of Control (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2006.02915

zbMATH Keywords

system identificationdynamical systemsneural networksdeep learning

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) System identification (93B30)

Related Items (6)

Sparse Bayesian deep learning for dynamic system identification ⋮ Learning nonlinear state-space models using autoencoders ⋮ Input-to-state stability for system identification with continuous-time Runge–Kutta neural networks ⋮ On the adaptation of recurrent neural networks for system identification ⋮ Deep subspace encoders for nonlinear system identification ⋮ Unnamed Item

Uses Software

• System Identification Toolbox
• CONTSID
• TensorFlow
• PyTorch
• Adam
• torchdiffeq
• PDE-Net

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Direct continuous-time approaches to system identification. Overview and benefits for practical applications
• 4SID: Subspace algorithms for the identification of combined deterministic-stochastic systems
• Comparative study of neural networks for dynamic nonlinear systems identification
• On the smoothness of nonlinear system identification
• PDE-Net 2.0: learning PDEs from data with a numeric-symbolic hybrid deep network
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• A proposal on machine learning via dynamical systems
• A flexible state-space model for learning nonlinear dynamical systems
• Runge-Kutta Methods with Minimum Error Bounds
• Learning Deep Architectures for AI
• Non-linear system identification using neural networks
• Convergence analysis of parametric identification methods
• Continuous-time model identification from sampled data: Implementation issues and performance evaluation
• Stable architectures for deep neural networks
• Finite-horizon integration for continuous-time identification: bias analysis and application to variable stiffness actuators
• Learning representations by back-propagating errors
• Special issue on ‘Applications of Continuous-Time Model Identification and Estimation’
• The advantages of directly identifying continuous-time transfer function models in practical applications

This page was built for publication: Continuous-time system identification with neural networks: model structures and fitting criteria