Search research articles

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

Network and Factor Structure of Depression and Anxiety Symptoms in Telemental Healthcare Patients From Bangladesh: Evidence for Precision Mental Healthcare.

Depression and anxiety·2026

Same author

Draft genome sequence of <i>Pseudomonas aeruginosa</i> SAU_MI_1F1 isolated from feces of cattle in Dhaka, Bangladesh.

Microbiology resource announcements·2026

Same author

Integrated in silico and in vitro assessment of Azadirachta indica leaf extract against multi-drug resistant Citrobacter koseri and Staphylococcus saprophyticus.

Scientific reports·2026

Same author

Early feasibility of telemedicine-based mental health wellbeing centers: an implementation study in district and sub-district health facilities in Bangladesh.

BMC health services research·2026

Same author

Tele-mental health for frail older adults in rural Bangladesh: a phenomenological study.

BMC psychology·2026

Same author

Draft genome sequence of <i>Salmonella enterica</i> subsp. <i>enterica</i> serovar Typhimurium SBI_US10_MRI_BD isolated from broiler chicken in Bangladesh.

Microbiology resource announcements·2026

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

Search research articles

相关实验视频

Updated: May 17, 2025

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents

Published on: May 3, 2012

在后端使用ChatGPT的自动化多参数神经架构发现框架.

Md Hafizur Rahman¹, Zafaryab Haider², Prabuddha Chakraborty²

¹Department of Electrical and Computer Engineering, University of Maine, Orono, ME, 04469, USA. md.hafizur.rahman@maine.edu.

Scientific reports

|May 15, 2025

概括

LEMONADE使用专家系统和大型语言模型自动设计边缘AI的神经网络架构. 这个框架简化了非专家的架构发现,在多个数据集上实现了最先进的结果.

科学领域:

人工智能的人工智能
机器学习机器学习
计算机科学计算机科学

背景情况:

设计高效的神经网络架构是复杂的,需要专业知识.
边缘人工智能 (AI) 带来了额外的挑战,包括功耗,模型大小和推断速度.

研究的目的:

引入一个新的框架,LEMONADE,用于自动神经网络架构的发现.
让非AI专家能够根据特定参数创建高效的模型,包括边缘AI约束.

主要方法:

开发了一个整合专家系统和在开放域知识上训练的大型语言模型 (LLM) 的框架.
使用了用户定义的参数,并考虑了边缘AI约束,而没有预定义的搜索空间.
使用CIFAR-10,CIFAR-100,ImageNet16-120,EuroSAT,疟疾寄生虫和IMDb等数据集验证了框架,使用ChatGPT-4o和Gemini-Pro作为LLMs.

主要成果:

生成的神经网络在CIFAR-10和CIFAR-100上实现了最先进的精度.
在ImageNet16-120数据集上展示了近乎最先进的性能.
成功创建了有效的神经网络,满足了像EuroSAT这样的数据集的各种边缘AI要求.

更多相关视频

Construction of an Improved Multi-Tetrode Hyperdrive for Large-Scale Neural Recording in Behaving Rats

Construction of an Improved Multi-Tetrode Hyperdrive for Large-Scale Neural Recording in Behaving Rats

Published on: May 9, 2018

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

相关实验视频

Last Updated: May 17, 2025

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents

Published on: May 3, 2012

Construction of an Improved Multi-Tetrode Hyperdrive for Large-Scale Neural Recording in Behaving Rats

Construction of an Improved Multi-Tetrode Hyperdrive for Large-Scale Neural Recording in Behaving Rats

Published on: May 9, 2018

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

结论:

LEMONADE为神经架构发现提供了一个可访问和有效的解决方案,特别是用于边缘AI应用程序.
该框架能够整合LLM和专家系统,使人工智能模型开发变得民主化.
在推进自动机器学习和高效的人工智能部署方面,LEMONADE显示出显著的前景.