Neurocomputing 469 (2022) 321–329
Contents lists available at Sciwith complexity
Nonlinear System identification
Long-term predictions controlled by a kernel-based strategy. The usefulness of the new approach is demonstrated through many
System stability examples, including important real benchmark problems taken from the system identification literature.
Nonparametric estimation Ó 2021 Elsevier . All rights reserved.
Deep networks
1. Introduction just assume that such map is smooth and will reconstruct it in a
nonparametric fashion. Hence, we will search for it in a very-
In many dynamical systems the relationship between the input high (possibly infinite-dimensional) space, then introducing regu-
and the output is described by a nonlinear function. Its estimation larization to control model complexity. Important approaches use
thus requires the introduction of a nonlinear model and the prob- kernels to encode in an implicit way smoothness information
lem to infer it from the available measurements is called
deep works文献 来自淘豆网m.daumloan.com转载请标明出处.
