Scholarlink

<p>Photonic computing offers an energy-efficient, high-bandwidth platform for artificial intelligence (AI) but currently faces scalability bottlenecks stemming from depth-dependent designs, linear optical structures, and intrinsic optical losses, along with high hardware and reconfiguration costs for multi-task processing. Here, we present a scaling paradigm that circumvents these limitations by expanding network width rather than depth, leveraging the intrinsic parallelism of photonics. We implement a scalable Photonic Mixture-of-Experts (PMoE) architecture, where parallel photonic cores function as expert networks. By dynamically routing inputs to these experts, the PMoE efficiently executes multi-task workloads without altering the physical optical weights. We fabricated a PMoE chip integrating three&#xa0;collaborative diffraction-based expert networks, featuring 18 parallel kernels within a compact intrinsic computational-core footprint of 0.067 mm<sup>2</sup>. Experimentally, the PMoE chip achieves multi-domain image classification with an average accuracy of 97.1%. While offering further scalability, this approach outperforms conventional optical networks and reduces digital parameter overhead by 67%. Our work underscores the scalability and efficiency of the PMoE architecture for next-generation large-scale photonic AI processors.</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

Photonic Mixture-of-Experts for scalable multi-task on-chip optical neural networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月3日

查看全文
  • Wencan Liu,
  • Zhenghang Zhang,
  • Peng Meng Chan,
  • Run Sun,
  • Yutong He,
  • Yuhao Wang,
  • Caihua Zhang,
  • Sigang Yang,
  • Tingzhao Fu,
  • Yuyao Huang,
  • Chaoran Huang,
  • Hongwei Chen

摘要

Photonic computing offers an energy-efficient, high-bandwidth platform for artificial intelligence (AI) but currently faces scalability bottlenecks stemming from depth-dependent designs, linear optical structures, and intrinsic optical losses, along with high hardware and reconfiguration costs for multi-task processing. Here, we present a scaling paradigm that circumvents these limitations by expanding network width rather than depth, leveraging the intrinsic parallelism of photonics. We implement a scalable Photonic Mixture-of-Experts (PMoE) architecture, where parallel photonic cores function as expert networks. By dynamically routing inputs to these experts, the PMoE efficiently executes multi-task workloads without altering the physical optical weights. We fabricated a PMoE chip integrating three collaborative diffraction-based expert networks, featuring 18 parallel kernels within a compact intrinsic computational-core footprint of 0.067 mm2. Experimentally, the PMoE chip achieves multi-domain image classification with an average accuracy of 97.1%. While offering further scalability, this approach outperforms conventional optical networks and reduces digital parameter overhead by 67%. Our work underscores the scalability and efficiency of the PMoE architecture for next-generation large-scale photonic AI processors.