Learning Nonlinear Heterogeneity in Physical Kolmogorov-Arnold Networks
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914310244007936 |
|---|---|
| author | Taglietti, Fabiana Pulici, Andrea Roxburgh, Maxwell Seguini, Gabriele Vidamour, Ian Menzel, Stephan Franco, Edoardo Laus, Michele Vasilaki, Eleni Perego, Michele Hayward, Thomas J. Fanciulli, Marco Gartside, Jack C. |
| author_facet | Taglietti, Fabiana Pulici, Andrea Roxburgh, Maxwell Seguini, Gabriele Vidamour, Ian Menzel, Stephan Franco, Edoardo Laus, Michele Vasilaki, Eleni Perego, Michele Hayward, Thomas J. Fanciulli, Marco Gartside, Jack C. |
| contents | Physical neural networks typically train linear synaptic weights while treating device nonlinearities as fixed. We show the opposite - by training the synaptic nonlinearity itself, as in Kolmogorov-Arnold Network (KAN) architectures, we yield markedly higher task performance per physical resource and improved performance-parameter scaling than conventional linear weight-based networks, demonstrating ability of KAN topologies to exploit reconfigurable nonlinear physical dynamics.
We experimentally realise physical KANs in silicon-on-insulator devices we term 'Synaptic Nonlinear Elements' (SYNEs), operating at room temperature, microampere currents, 2 MHz speeds and ~750 fJ per nonlinear operation, with no observed degradation over 10^13 measurements and months-long timescales.
We demonstrate nonlinear function regression, classification, and prediction of Li-Ion battery dynamics from noisy real-world multi-sensor data. Physical KANs outperform equivalently-parameterised software multilayer perceptron networks across all tasks, with up to two orders of magnitude fewer parameters, and two orders of magnitude fewer devices than linear weight based physical networks. These results establish learned physical nonlinearity as a hardware-native computational primitive for compact and efficient learning systems, and SYNE devices as effective substrates for heterogenous nonlinear computing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_15340 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning Nonlinear Heterogeneity in Physical Kolmogorov-Arnold Networks Taglietti, Fabiana Pulici, Andrea Roxburgh, Maxwell Seguini, Gabriele Vidamour, Ian Menzel, Stephan Franco, Edoardo Laus, Michele Vasilaki, Eleni Perego, Michele Hayward, Thomas J. Fanciulli, Marco Gartside, Jack C. Disordered Systems and Neural Networks Mesoscale and Nanoscale Physics Machine Learning Adaptation and Self-Organizing Systems Applied Physics Physical neural networks typically train linear synaptic weights while treating device nonlinearities as fixed. We show the opposite - by training the synaptic nonlinearity itself, as in Kolmogorov-Arnold Network (KAN) architectures, we yield markedly higher task performance per physical resource and improved performance-parameter scaling than conventional linear weight-based networks, demonstrating ability of KAN topologies to exploit reconfigurable nonlinear physical dynamics. We experimentally realise physical KANs in silicon-on-insulator devices we term 'Synaptic Nonlinear Elements' (SYNEs), operating at room temperature, microampere currents, 2 MHz speeds and ~750 fJ per nonlinear operation, with no observed degradation over 10^13 measurements and months-long timescales. We demonstrate nonlinear function regression, classification, and prediction of Li-Ion battery dynamics from noisy real-world multi-sensor data. Physical KANs outperform equivalently-parameterised software multilayer perceptron networks across all tasks, with up to two orders of magnitude fewer parameters, and two orders of magnitude fewer devices than linear weight based physical networks. These results establish learned physical nonlinearity as a hardware-native computational primitive for compact and efficient learning systems, and SYNE devices as effective substrates for heterogenous nonlinear computing. |
| title | Learning Nonlinear Heterogeneity in Physical Kolmogorov-Arnold Networks |
| topic | Disordered Systems and Neural Networks Mesoscale and Nanoscale Physics Machine Learning Adaptation and Self-Organizing Systems Applied Physics |
| url | https://arxiv.org/abs/2601.15340 |