State-Space Kolmogorov Arnold Networks for Interpretable Nonlinear System Identification
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908417221722112 |
|---|---|
| author | Cruz, Gonçalo Granjal Renczes, Balazs Runacres, Mark C Decuyper, Jan |
| author_facet | Cruz, Gonçalo Granjal Renczes, Balazs Runacres, Mark C Decuyper, Jan |
| contents | While accurate, black-box system identification models lack interpretability of the underlying system dynamics. This paper proposes State-Space Kolmogorov-Arnold Networks (SS-KAN) to address this challenge by integrating Kolmogorov-Arnold Networks within a state-space framework. The proposed model is validated on two benchmark systems: the Silverbox and the Wiener-Hammerstein benchmarks. Results show that SS-KAN provides enhanced interpretability due to sparsity-promoting regularization and the direct visualization of its learned univariate functions, which reveal system nonlinearities at the cost of accuracy when compared to state-of-the-art black-box models, highlighting SS-KAN as a promising approach for interpretable nonlinear system identification, balancing accuracy and interpretability of nonlinear system dynamics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_16392 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | State-Space Kolmogorov Arnold Networks for Interpretable Nonlinear System Identification Cruz, Gonçalo Granjal Renczes, Balazs Runacres, Mark C Decuyper, Jan Machine Learning Systems and Control While accurate, black-box system identification models lack interpretability of the underlying system dynamics. This paper proposes State-Space Kolmogorov-Arnold Networks (SS-KAN) to address this challenge by integrating Kolmogorov-Arnold Networks within a state-space framework. The proposed model is validated on two benchmark systems: the Silverbox and the Wiener-Hammerstein benchmarks. Results show that SS-KAN provides enhanced interpretability due to sparsity-promoting regularization and the direct visualization of its learned univariate functions, which reveal system nonlinearities at the cost of accuracy when compared to state-of-the-art black-box models, highlighting SS-KAN as a promising approach for interpretable nonlinear system identification, balancing accuracy and interpretability of nonlinear system dynamics. |
| title | State-Space Kolmogorov Arnold Networks for Interpretable Nonlinear System Identification |
| topic | Machine Learning Systems and Control |
| url | https://arxiv.org/abs/2506.16392 |