Scholarlink

<p>Photonic neural networks promise inference at the speed of light, yet most architectures combine linear optical meshes with electronic nonlinearities, reintroducing optical-electrical-optical bottlenecks. Kolmogorov-Arnold networks place trainable nonlinear functions on network edges, concentrating expressivity into a few structured modules. Each edge here is a single module built from a Mach-Zehnder interferometer, a semiconductor optical amplifier, and variable optical attenuators, giving a four-parameter transfer function set by gain saturation and interferometric mixing. A four-module network attains 94.3% accuracy (±3.9% s.d. over ten seeds) on nonlinear classification, and a seven-module network reaches <i>R</i><sup>2</sup>&#xa0;=&#xa0;0.986&#xa0;±&#xa0;0.015 on six-input regression, remaining robust to 6-bit inputs and 14 dB signal-to-noise ratio. Here, we show that a fully differentiable physics model enables end-to-end optimization of these standard telecom modules, giving a practical route from simulation toward experimental demonstration of photonic Kolmogorov-Arnold networks.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

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

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月21日

查看全文
  • Luca Nogueira Calçado,
  • Sergei K. Turitsyn,
  • Egor Manuylovich

摘要

Photonic neural networks promise inference at the speed of light, yet most architectures combine linear optical meshes with electronic nonlinearities, reintroducing optical-electrical-optical bottlenecks. Kolmogorov-Arnold networks place trainable nonlinear functions on network edges, concentrating expressivity into a few structured modules. Each edge here is a single module built from a Mach-Zehnder interferometer, a semiconductor optical amplifier, and variable optical attenuators, giving a four-parameter transfer function set by gain saturation and interferometric mixing. A four-module network attains 94.3% accuracy (±3.9% s.d. over ten seeds) on nonlinear classification, and a seven-module network reaches R2 = 0.986 ± 0.015 on six-input regression, remaining robust to 6-bit inputs and 14 dB signal-to-noise ratio. Here, we show that a fully differentiable physics model enables end-to-end optimization of these standard telecom modules, giving a practical route from simulation toward experimental demonstration of photonic Kolmogorov-Arnold networks.