Deep neural network based adaptive learning for switched systems
From MaRDI portal
Abstract: In this paper, we present a deep neural network based adaptive learning (DNN-AL) approach for switched systems. Currently, deep neural network based methods are actively developed for learning governing equations in unknown dynamic systems, but their efficiency can degenerate for switching systems, where structural changes exist at discrete time instants. In this new DNN-AL strategy, observed datasets are adaptively decomposed into subsets, such that no structural changes within each subset. During the adaptive procedures, DNNs are hierarchically constructed, and unknown switching time instants are gradually identified. Especially, network parameters at previous iteration steps are reused to initialize networks for the later iteration steps, which gives efficient training procedures for the DNNs. For the DNNs obtained through our DNN-AL, bounds of the prediction error are established. Numerical studies are conducted to demonstrate the efficiency of DNN-AL.
Recommendations
- Online learning of data-driven controllers for unknown switched linear systems
- Data-driven learning of nonautonomous systems
- Data driven governing equations approximation using deep neural networks
- scientific article; zbMATH DE number 7338324
- Adaptive neural network control scheme of switched systems with input saturation
Cites work
- A proposal on machine learning via dynamical systems
- An eigensystem realization algorithm for modal parameter identification and model reduction
- An introduction to hybrid dynamical systems
- Automatic Selection of Methods for Solving Stiff and Nonstiff Systems of Ordinary Differential Equations
- Complex systems analysis of series of blackouts: Cascading failure, critical points, and self-organization
- Data driven governing equations approximation using deep neural networks
- Data-driven learning of nonautonomous systems
- Deep learning
- Deep learning nonlinear multiscale dynamic problems using Koopman operator
- Deep learning of dynamics and signal-noise decomposition with time-stepping constraints
- Discontinuous dynamical systems
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Equation-free, coarse-grained multiscale computation: enabling microscopic simulators to perform system-level analysis
- Extracting Sparse High-Dimensional Dynamics from Limited Data
- scientific article; zbMATH DE number 1405266 (Why is no real title available?)
- Identification of observer/Kalman filter Markov parameters - Theory and experiments
- Kalman filters for non-linear systems: a comparison of performance
- Machine learning for prediction with missing dynamics
- Model selection for hybrid dynamical systems via sparse regression
- Modern Koopman theory for dynamical systems
- Optimization methods for large-scale machine learning
- Ordinary differential equations and dynamical systems
- Revealing hidden dynamics from time-series data by ODENet
- Robust data-driven discovery of governing physical laws with error bars
- Seasonally forced disease dynamics explored as switching between attractors
- Switching in systems and control
- The Dynamics of Legged Locomotion: Models, Analyses, and Challenges
Cited in
(3)
This page was built for publication: Deep neural network based adaptive learning for switched systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6172098)
