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

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

Effect of Steam Curing Regimes on Mechanical Performance, Shrinkage and Microstructure of Fly Ash-Slag-Desulfurization Gypsum Cementitious Materials.

Materials (Basel, Switzerland)·2026
Same author

Association between IoT-driven dynamic pulmonary function monitoring and quality of life in community-dwelling patients with chronic obstructive pulmonary disease: a 2-year longitudinal study.

Frontiers in medicine·2026
Same author

Palmitoyltransferase DHHC7 mediates protein palmitoylation and is essential for sperm function through the modulation of [Ca<sup>2+</sup>]<sub>i</sub> and ROS signaling.

Frontiers in cell and developmental biology·2026
Same author

Dissecting the immune-metabolic axis in appendicitis: A Mendelian randomization mediation study.

Medicine·2026
Same author

Metagenomic and Metabolomic Analysis of Intestinal Excrement Differences Between Natural Hatching and Artificial Peeling out of the Shell in <i>Nipponia nippon</i>.

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

Transcriptomic and Metabolomic Profiling Reveals the Antiproliferative Mechanism of Goose Serum and Plasma in SW1990 Cells.

Biology·2026

相关实验视频

Updated: Jan 11, 2026

Biomolecular Imaging of Cellular Uptake of Nanoparticles using Multimodal Nonlinear Optical Microscopy
07:13

Biomolecular Imaging of Cellular Uptake of Nanoparticles using Multimodal Nonlinear Optical Microscopy

Published on: May 16, 2022

2.3K

预测多式光学响应超快的基于等离子体的功能性通用近似定理,具有指数级数据效率.

Yulu Qin, Haoyang Cheng, Haixia Zheng

    Optics express
    |November 11, 2025
    PubMed
    概括

    这项研究引入了一个新的机器学习框架,功能通用近似 (FUA),用于纳米光子学. 与传统神经网络相比,FUA有效地模拟复杂的光学响应,使用的数据少于传统神经网络,提高了反向设计能力.

    科学领域:

    • 纳米光子学 纳米光子学
    • 计算电磁学的计算.
    • 机器学习是机器学习.

    背景情况:

    • 纳米光子学的传统机器学习方法经常与功能数据扎,由于无法捕获固有的数据结构,因此需要大数据集.
    • 神经网络 (NN) 方法通常将功能数据模型为高维向量,忽视了流性和连续性,这限制了它们的效率和稳定性.

    研究的目的:

    • 通过利用功能数据分析 (FDA) 来开发纳米光子反向设计的数据效率模型框架.
    • 解决传统的基于NN的方法在捕捉纳米光子结构-光学响应关系的功能性质的局限性.

    主要方法:

    • 提出了一个基于功能通用近似 (FUA) 定理的新框架,整合了FDA的原则.
    • 显式建模功能结构以学习非线性函数-在-尺度映射.
    • 在纳米环磁盘二极管模型上验证了FUA方法.

    主要成果:

    • 对吸收 (R 2=0.86),散射 (R 2=0.84),近场光谱 (R 2=0.96) 和时间解析电场 (R 2=0.98) 实现了高预测准确度.
    • 证明了卓越的数据效率,只需要300个训练样本.
    • 与标准NN模型相比,展示了增强的稳定性和概括能力.

    更多相关视频

    Determination of the Excitation and Coupling Rates Between Light Emitters and Surface Plasmon Polaritons
    07:39

    Determination of the Excitation and Coupling Rates Between Light Emitters and Surface Plasmon Polaritons

    Published on: July 21, 2018

    7.2K
    Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping
    09:43

    Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping

    Published on: March 20, 2017

    10.3K

    相关实验视频

    Last Updated: Jan 11, 2026

    Biomolecular Imaging of Cellular Uptake of Nanoparticles using Multimodal Nonlinear Optical Microscopy
    07:13

    Biomolecular Imaging of Cellular Uptake of Nanoparticles using Multimodal Nonlinear Optical Microscopy

    Published on: May 16, 2022

    2.3K
    Determination of the Excitation and Coupling Rates Between Light Emitters and Surface Plasmon Polaritons
    07:39

    Determination of the Excitation and Coupling Rates Between Light Emitters and Surface Plasmon Polaritons

    Published on: July 21, 2018

    7.2K
    Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping
    09:43

    Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping

    Published on: March 20, 2017

    10.3K

    结论:

    • FUA框架为建模纳米光子系统提供了强大且数据效率高的替代方案.
    • 这种方法在纳米光子学中实现高效的反向设计方面显著有前途,特别是在有限的数据条件下.
    • FUA有效地捕捉了结构-属性关系的功能性质,在数据效率和概括方面表现优于传统的NN方法.