发表机构
Tongji University; King Abdullah University of Science and Technology (KAUST); Tsinghua University; EPFL（同济大学; 阿卜杜拉国王科技大学; 清华大学; 洛桑联邦理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出自整流非线性光学神经处理单元（ONNPU），实现全光非线性激活，兼容深度学习生态，在多项任务中表现优异，推动实用光学计算发展。
AI 中文摘要
人工智能（AI）和神经网络已在众多学科中推动了突破性创新。光学计算在后摩尔时代有望提供前所未有的速度和能效；然而，利用全光方法实现高效、实用的非线性激活仍然是一个挑战。在此，我们提出了一种光学非线性神经处理单元（ONNPU），通过自整流机制实现全光非线性激活。ONNPU架构完美模仿了数字神经网络的结构，使其能够与成熟的深度学习生态系统无缝集成。我们在涵盖决策、回归和生成的九个不同任务上对ONNPU进行了基准测试，包括在MNIST上达到98.07%的准确率，在Fashion-MNIST上达到93.54%的准确率。当集成到拥有2.01亿参数的视觉Transformer中时，ONNPU在完整的ImageNet分类（1000个类别）上实现了82.4%的top-1准确率；当集成到拥有1.17亿参数的仅解码器Transformer中时，ONNPU实现了短篇故事生成，性能优于GPT-2。通过处理更复杂和多样化的深度学习任务，ONNPU为实用的光学机器智能铺平了道路，释放了高性能光学计算的巨大潜力。
英文摘要
Artificial intelligence (AI) and neural networks have driven groundbreaking innovations across numerous disciplines. Optical computing offers the promise of unprecedented speed and energy efficiency in the post-Moore era; however, achieving efficient, practical nonlinear activation using all-optical approaches remains a challenge. Here, we present an optical nonlinear neural processing unit (ONNPU) that implements all-optical nonlinear activation through a self-rectification mechanism. The ONNPU architecture perfectly imitates the structure of digital neural networks, enabling seamless integration with the established deep learning ecosystem. We benchmark ONNPU across nine diverse tasks spanning decision, regression and generation, including accuracies of 98.07% on MNIST and 93.54% on Fashion-MNIST. When integrated into a 201-million-parameter Vision Transformer, ONNPU achieves 82.4% top-1 accuracy on full ImageNet classification (1,000 categories); when integrated into a 117-million-parameter decoder-only Transformer, ONNPU enables short-form story generation that outperforms GPT-2. By addressing more complex and diverse deep learning tasks, ONNPU paves the way toward practical optical machine intelligence, unleashing significant potential for high-performance optical computing.
Comments33 pages, 6 figures, 1 table
