Decoupling P-NARX models using filtered CPD

Decuyper, Jan; Westwick, David T.; Karami, Kiana; Schoukens, Johan

doi:10.1016/j.ifacol.2021.08.436

Cited by 2 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar test set performance was obtained for the competing method of [12], yielding a value of 1.78% given the same settings, i.e. r = 4, n u = 1 and n y = 3 [24], and a value of 0.9% in case of a slightly larger model with r = 4, n u = 2 and n y = 3 [12].…”

Section: Decoupled P-narx Modelsupporting

confidence: 63%

See 1 more Smart Citation

Decoupling multivariate functions using a nonparametric filtered tensor decomposition

Decuyper,

Tiels,

Weiland

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Multivariate functions emerge naturally in a wide variety of data-driven models. Popular choices are expressions in the form of basis expansions or neural networks. While highly effective, the resulting functions tend to be hard to interpret, in part because of the large number of required parameters. Decoupling techniques aim at providing an alternative representation of the nonlinearity. The so-called decoupled form is often a more efficient parameterisation of the relationship while being highly structured, favouring interpretability. In this work two new algorithms, based on filtered tensor decompositions of first order derivative information are introduced. The method returns nonparametric estimates of smooth decoupled functions. Direct applications are found in, i.a. the fields of nonlinear system identification and machine learning.

show abstract

Section: Decoupled P-narx Modelsupporting

confidence: 63%

“…This turns out to be an important class of functions, frequently encountered in practice. The examples in [6,24,25,7] show that nonlinearities found in dynamical models can often be brought back to a low number of underlying characteristic univariate functions. This can lead to insight into the system's behaviour.…”

Section: Polynomial Reductionmentioning

confidence: 99%