Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

Photoreceptors and Visual Pathways01:22

Photoreceptors and Visual Pathways

8.7K
At the molecular level, visual signals trigger transformations in photopigment molecules, resulting in changes in the photoreceptor cell's membrane potential. The photon's energy level is denoted by its wavelength, with each specific wavelength of visible light associated with a distinct color. The spectral range of visible light, classified as electromagnetic radiation, spans from 380 to 720 nm. Electromagnetic radiation wavelengths exceeding 720 nm fall under the infrared category,...
8.7K

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

A review of the application of novel intervertebral disc diagnostic technologies integrated with artificial intelligence in medical imaging.

Digital health·2026
Same author

Cuproptosis-immunity crosstalk informs strategy to overcome immunotherapy resistance.

Cell·2026
Same author

Multimorbidity Patterns and Cognitive Transitions Among the Elderly in China: The Longitudinal Evidence From CLHLS.

Asia-Pacific journal of public health·2026
Same author

A skin-conformal rigid-in-soft array-based imaging system.

Nature communications·2026
Same author

An Autocatalytic Molecular Sensor Enables Rapid, Sensitive, and One-Pot Detection of Nucleic Acids.

ACS nano·2026
Same author

Effects of Traditional Chinese Herbal Extracts on Tear Staining, Iron Status, Immune Function, and Antioxidant Capacity in Dogs.

Animals : an open access journal from MDPI·2026
Same journal

Demonstration of a quantum C-NOT gate in a time-multiplexed fully reconfigurable photonic processor.

Nature communications·2026
Same journal

Nonlinear quantum light source with van der Waals ferroelectric NbOX<sub>2</sub> (X = Br, I).

Nature communications·2026
Same journal

Antagonistic histone H2A variants and autonomous heterochromatin formation shape epigenomic patterns in Arabidopsis.

Nature communications·2026
Same journal

The long tail of nitrate pollution in groundwater challenges governance of global water quality.

Nature communications·2026
Same journal

Select microbial metabolites promote tau aggregation in a murine tauopathy model.

Nature communications·2026
Same journal

Warming climate has lengthened global intense tropical cyclone seasons.

Nature communications·2026
查看所有相关文章

相关实验视频

Updated: Jan 10, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

999

一百层光子深度学习

Tiankuang Zhou1,2, Yizhou Jiang1,2, Zhihao Xu1,2

  • 1Department of Electronic Engineering, Tsinghua University, Beijing, China.

Nature communications
|November 24, 2025
PubMed
概括
此摘要是机器生成的。

研究人员开发了一种新的单层光子计算 (SLiM) 芯片,可以克服光学神经网络中的错误积累. 这一突破使得先进的人工智能应用程序的数百个层的深度学习模型成为可能.

更多相关视频

Lensless Fluorescent Microscopy on a Chip
11:23

Lensless Fluorescent Microscopy on a Chip

Published on: August 17, 2011

18.1K

相关实验视频

Last Updated: Jan 10, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

999
Lensless Fluorescent Microscopy on a Chip
11:23

Lensless Fluorescent Microscopy on a Chip

Published on: August 17, 2011

18.1K

科学领域:

  • 光子学和人工智能的人工智能
  • 集成光子电路 集成光子电路
  • 深度学习 硬件 硬件

背景情况:

  • 由于错误积累,光子计算的模拟性限制了神经网络深度.
  • 目前的光学神经网络仅限于大约十个层,阻碍了像大型语言模型 (LLM) 这样的高级AI功能.
  • 光学系统中的传播冗余在很大程度上导致累积的错误.

研究的目的:

  • 为了克服光学神经网络的深度限制.
  • 为深度学习开发一个耐错误的光子计算芯片.
  • 在模拟硬件上实现高能效的先进AI模型.

主要方法:

  • 引入了芯片上的扰动来解计算相关性并消除传播冗余.
  • 开发了一个单层光子计算 (SLiM) 芯片架构.
  • 使用SLiM芯片进行实验,构建了包括LLM在内的深度神经网络.

主要成果:

  • 该SLiM芯片展示了容错性,限制了超过200层的错误率.
  • 扩展空间深度从毫米到百米尺度,促进3D芯片集群.
  • 实现了与100层图像分类网络的理想模拟和10GHz数据速率的文本和图像生成的LLM相似的性能.

结论:

  • 耐错误的SLiM芯片消除了光学神经网络的深度限制.
  • 这一进步使得在高效的模拟计算硬件上实现最先进的深度学习模型.
  • 为更强大,更节能的人工智能系统铺平了道路.