Search research articles

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 Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

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

A Liposomal Nanovaccine Reprograms Post-Ablation Macrophages to Sustain Antitumor Immunity in Hepatocellular Carcinoma.

Acta biomaterialia·2026

Same author

Integration of transcriptome profiling to identify key genes involved in the interplay between oxidative stress and mitophagy in major depressive disorder, followed by multidimensional phenotypic validation.

Frontiers in psychiatry·2026

Same author

[Screening of novel uric acid-lowering probiotics and evaluation of their probiotic properties].

Sheng wu gong cheng xue bao = Chinese journal of biotechnology·2026

Same author

Molecular mechanisms of KMT2C alterations in gastrointestinal cancers: enhancer network destabilization, lineage plasticity, and clinical translation.

Frontiers in immunology·2026

Same author

Reflective-path-enabled folded AWG on thin-film lithium niobate with 20 GHz fine channel spacing.

Optics express·2026

Same author

The coping strategies and needs of caregiving burnout among family caregivers of elderly stroke survivors.

Geriatric nursing (New York, N.Y.)·2026

Search research articles

相关实验视频

Updated: Sep 11, 2025

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Published on: March 2, 2011

基于DML和DFB-SA激光芯片的光子尖端神经网络用于模式分类.

Xintao Zeng, Shuiying Xiang, Yanan Han

|August 13, 2025

概括

这项研究介绍了一种紧的光子尖端神经网络 (PSNN),使用直接调制的激光和可和吸收激光. 新设计实现了高能效和强大的性能,用于AI加速.

科学领域:

光子学是一个光子学.
人工智能的人工智能
计算机科学计算机科学

背景情况:

神经形态光子计算为先进的人工智能加速器提供了潜力,因为它的速度和低能耗.
当前的神经形态光子系统面临着由于大型离散组件而导致可扩展性的限制.

研究的目的:

为光子尖端神经网络 (PSNNs) 提出一种新的,紧的,可扩展的架构.
为了证明使用直接调制的激光和分布式反激光与和吸收器 (DML-DFB-SA) 的可行性,用于PSNNs.
在神经形态光子计算中增强能源效率和集成能力.

主要方法:

开发了一个PSNN架构,集成直接调制激光器 (DML) 作为光源和电光转换器,以及带有和吸收器 (DFB-SA) 的分布式反激光器作为尖端神经元.
采用时间复合的尖峰编码方案,使单个神经元能够模拟多个功能.
在酸 (InP) 基板上制造的DML和DFB-SA激光芯片,用于潜在的大规模集成.

主要成果:

实现了值得注意的0.625 pJ/MAC的能量效率.
在短距离传输中证明了对DML的声效应的强大性能.
在使用拟议的DML-DFB-SA系统与时间复合编码的MNIST数据集上实现了94%的识别准确度.

更多相关视频

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

Published on: March 25, 2014

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

Published on: September 5, 2012

相关实验视频

Last Updated: Sep 11, 2025

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Published on: March 2, 2011

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

Published on: March 25, 2014

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

Published on: September 5, 2012

结论:

DML-DFB-SA架构为神经形态光子计算提供了一个紧,节能和可扩展的解决方案.
这种方法促进了用于下一代AI加速器的大规模集成PSNN芯片的开发.
该系统的稳定性和高性能使其成为实际PSNN应用的有希望的候选者.