|国家预印本平台
首页|A scalable and programmable optical neural network in a time-synthetic dimension

A scalable and programmable optical neural network in a time-synthetic dimension

A scalable and programmable optical neural network in a time-synthetic dimension

来源:Arxiv_logoArxiv
英文摘要

Programmable optical neural networks (ONNs) can offer high-throughput and energy-efficient solutions for accelerating artificial intelligence (AI) computing. However, existing ONN architectures, typically based on cascaded unitary transformations such as Mach-Zehnder interferometer meshes, face inherent scalability limitations due to spatial encoding, which causes optical components and system complexity to scale quadratically with network size. A promising solution to this challenge is the use of synthetic dimensions to enhance scalability, though experimental demonstration has remained scarce. Here, we present the first experimental demonstration of an all-optical, highly scalable, programmable ONN operating in a time-synthetic dimension. By implementing a time-cycle computation paradigm analogous to gate cycling in conventional spatial photonic circuits, our approach achieves a gate count surpassing that of state-of-the-art programmable photonic processors. Unlike conventional ONN architectures that rely on real-space wave interferences, our framework exploits time-reflection and time-refraction to perform computations, fundamentally eliminating backscattering errors through causality constraints. To bridge the gap between simulation and reality, we introduce an in-situ training framework that dynamically adapts to experimental errors, achieving performance exceeding traditional in silico learning paradigms. Our synthetic-dimension-based approach provides a compact, scalable, backscattering-free, and programmable neuromorphic computing architecture, advancing the potential for next-generation photonic AI systems.

Yihao Yang、Bei Wu、Yudong Ren、Rui Zhao、Haiyao Luo、Fujia Chen、Li Zhang、Lu Zhang、Hongsheng Chen

光电子技术

Yihao Yang,Bei Wu,Yudong Ren,Rui Zhao,Haiyao Luo,Fujia Chen,Li Zhang,Lu Zhang,Hongsheng Chen.A scalable and programmable optical neural network in a time-synthetic dimension[EB/OL].(2025-07-04)[2025-07-16].https://arxiv.org/abs/2507.02297.点此复制

评论