🤖 AI Summary
Current photonic hardware struggles to support general-purpose artificial intelligence computing due to the lack of scalable high-order nonlinear capabilities. This work proposes a hybrid optoelectronic architecture that leverages optical random Fourier feature mapping to transform high-dimensional nonlinear function approximation into linear operations, thereby enabling universal nonlinear processing on purely linear silicon photonic chips—without requiring complex nonlinear materials or active components. The approach combines scalability with high throughput and is experimentally validated through efficient implementations of tenth-order Legendre polynomials, special functions such as Voigt, Fermi-Dirac, and Fresnel profiles, neural network activation functions, two-dimensional nonlinear mappings, and a 10-dimensional softmax layer. This study thus demonstrates, for the first time, universal nonlinear computation using only linear photonic circuits.
📝 Abstract
Photonic computing has emerged as a promising platform for accelerating artificial intelligence workloads by enabling low-latency and energy-efficient linear operations such as vector-matrix multiplication. However, scalable on-chip high-order nonlinear processing remains challenging, limiting the functional versatility of current photonic hardware. Here, we present an optoelectronic approach for approximating high-order and high-dimensional nonlinear functions. The key to this approach lies in optical random Fourier feature mapping, which transforms nonlinear function evaluation into an equivalent linear computation. This approach enables nonlinear computing within a linear photonic framework, eliminating the need for complex optical nonlinear or active materials while preserving scalability and computational throughput in a simple silicon photonic circuit. We experimentally demonstrate a broad class of nonlinear functions, including tenth-order Legendre polynomials, computationally demanding special functions (Voigt, Fermi-Dirac, and Fresnel), neural-network activation functions, two-dimensional nonlinear functions, and a 10-dimensional softmax layer. This work establishes a general and scalable strategy for nonlinear computing in photonic integrated hardware and opens a pathway toward fully functional optical accelerators for next-generation computing systems.