System identification: A learning theory approach
The problem of system identification is formulated as a problem in statistical learning theory. lt is proved that if a particular type of uniform convergence result holds, then the classical approach of choosing the current model to minimize the error on the observed data can converge to an optimal model within the specified class. The reformulation of the problem leads to finite time estimates for the rate at which the estimated model converges to the optimal model. For exponentially stable systems it is proved that the desired uniform convergence holds so that the learning theory approach is applicable. For the entire collection see [Zbl 1007.00075].
- On convergence proofs in system identification -- a general principle using ideas from learning theory
- A Learning Theory Approach to System Identification and Stochastic Adaptive Control
- Statistical learning theory, model identification and system information content
- Learning theory approach to a system identification problem involving atomic norm
- Learning dynamical systems in a stationary environment
- On convergence proofs in system identification -- a general principle using ideas from learning theory
- An adaptive learning approach to the identification of structural and mechanical systems
- Learning dynamical systems in a stationary environment
- System laws and identification theory
- Learning processes based on incomplete identification and information generation
- System models or learning machines?
- Learning theory and robust control
- Neural ARX models and PAC learning
- Statistical learning theory, model identification and system information content
- scientific article; zbMATH DE number 3850963 (Why is no real title available?)
- scientific article; zbMATH DE number 3852929 (Why is no real title available?)
- scientific article; zbMATH DE number 3924653 (Why is no real title available?)
- scientific article; zbMATH DE number 4053057 (Why is no real title available?)
- Inverse passive learning of an input-output-map through update-spline-smoothing
- System identification using multilayer differential neural networks: a new result
- Time complexity and model complexity of fast identification of continuous-time LTI systems
- On system identification of complex systems from finite data
- Gradient descent learns linear dynamical systems
- Learning Complexity Dimensions for a Continuous-Time Control System
- scientific article; zbMATH DE number 867220 (Why is no real title available?)
- Learning identification: a least squares algorithm and its repetitive consistency
- Identification of linear dynamical systems and machine learning
- A Learning Theory Approach to System Identification and Stochastic Adaptive Control
- A shift in paradigm for system identification
- Using statistical learning theory to rationalize system model identification and validation. I: Mathematical foundations
- scientific article; zbMATH DE number 3394006 (Why is no real title available?)
- Explicit convergence relations for a class of discrete LTV systems and its application to performance analysis of deterministic learning
- Signed-perturbed sums estimation of ARX systems: exact coverage and strong consistency
- Learning theory approach to a system identification problem involving atomic norm
This page was built for publication: System identification: A learning theory approach
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4437830)
