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

TL;DR

This paper introduces small-scale photonic neural networks built entirely with standard telecom components, achieving high accuracy in classification and regression tasks with minimal hardware and robust performance.

Contribution

The work demonstrates the feasibility of implementing nonlinear photonic neural networks using standard telecom modules, enabling practical and efficient optical computing.

Findings

Achieved 94.3% accuracy on nonlinear classification benchmarks.

Attained R^2 = 0.986 on six-input regression tasks.

Maintained high performance with 6-bit input resolution and 14 dB SNR.

Abstract

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.