Small-scale photonic Kolmogorov-Arnold networks using standard telecom nonlinear modules

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Calçado, Luca Nogueira, Turitsyn, Sergei K., Manuylovich, Egor
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916022221537280
author Calçado, Luca Nogueira
Turitsyn, Sergei K.
Manuylovich, Egor
author_facet Calçado, Luca Nogueira
Turitsyn, Sergei K.
Manuylovich, Egor
contents Photonic neural networks promise ultrafast inference, yet most architectures rely on linear optical meshes with electronic nonlinearities, reintroducing optical-electrical-optical bottlenecks. Here we introduce small-scale photonic Kolmogorov-Arnold networks (SSP-KANs) implemented entirely with standard telecommunications components. Each network edge employs a trainable nonlinear module composed of a Mach-Zehnder interferometer, semiconductor optical amplifier, and variable optical attenuators, providing a four-parameter transfer function derived from gain saturation and interferometric mixing. Despite the constrained functional form of these optical nonlinearities, SSP-KANs comprising only a few optical modules achieve strong nonlinear inference performance across classification, regression, and image recognition tasks, approaching software baselines with significantly fewer parameters. A four-module network achieves $94.3$\% (IQR: $90.3$--$97.4$\%, 10~seeds) accuracy on nonlinear classification benchmarks; a seven-module network attains $R^2 = 0.986 \pm 0.015$ on six-input regression. Performance remains robust under realistic hardware impairments, maintaining high accuracy down to 6-bit input resolution and 14 dB signal-to-noise ratio. By using a fully differentiable physics model for end-to-end optimisation of optical parameters, this work establishes a practical pathway from simulation to experimental demonstration of photonic KANs using commodity telecom hardware.
format Preprint
id arxiv_https___arxiv_org_abs_2604_08432
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Small-scale photonic Kolmogorov-Arnold networks using standard telecom nonlinear modules
Calçado, Luca Nogueira
Turitsyn, Sergei K.
Manuylovich, Egor
Optics
Artificial Intelligence
Photonic neural networks promise ultrafast inference, yet most architectures rely on linear optical meshes with electronic nonlinearities, reintroducing optical-electrical-optical bottlenecks. Here we introduce small-scale photonic Kolmogorov-Arnold networks (SSP-KANs) implemented entirely with standard telecommunications components. Each network edge employs a trainable nonlinear module composed of a Mach-Zehnder interferometer, semiconductor optical amplifier, and variable optical attenuators, providing a four-parameter transfer function derived from gain saturation and interferometric mixing. Despite the constrained functional form of these optical nonlinearities, SSP-KANs comprising only a few optical modules achieve strong nonlinear inference performance across classification, regression, and image recognition tasks, approaching software baselines with significantly fewer parameters. A four-module network achieves $94.3$\% (IQR: $90.3$--$97.4$\%, 10~seeds) accuracy on nonlinear classification benchmarks; a seven-module network attains $R^2 = 0.986 \pm 0.015$ on six-input regression. Performance remains robust under realistic hardware impairments, maintaining high accuracy down to 6-bit input resolution and 14 dB signal-to-noise ratio. By using a fully differentiable physics model for end-to-end optimisation of optical parameters, this work establishes a practical pathway from simulation to experimental demonstration of photonic KANs using commodity telecom hardware.
title Small-scale photonic Kolmogorov-Arnold networks using standard telecom nonlinear modules
topic Optics
Artificial Intelligence
url https://arxiv.org/abs/2604.08432