CAREER: An Embodied Intelligence Approach to Neural Architecture Search
职业:神经架构搜索的具身智能方法
基本信息
- 批准号:2239691
- 负责人:
- 金额:$ 54.99万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Standard Grant
- 财政年份:2023
- 资助国家:美国
- 起止时间:2023-09-01 至 2028-08-31
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
Recent advances in Machine Learning (ML) models like deep neural networks have shown immense promise to solve problems in a wide variety of fields, ranging from health and wellness to environmental science to national defense. Yet the practical use of these methods currently requires a great deal of ML-training and experience to implement due to their sensitivity to nonintuitive and complex configuration settings like the size and shape of deep neural networks. The subfield of Automated Machine Learning (AutoML) seeks to help reduce the barriers to entry by creating models and pipelines which automatically self-configure based on the needs of a given problem. Within AutoML, Neural Architecture Search (NAS) aims to automatically find the ideal structure of a deep neural network. The process of using ML to find the optimal shape and form of a deep neural network is roughly analogous to the well-studied evolutionary and developmental processes that create shape and form in biological creatures, or that automatically find the shape and form of robots in the field of Evolutionary Robotics. Despite the analogies between these subfields, and the outsized impact that work in AutoML may have on our ability to Harness the Data Revolution and make practical impacts across a wide variety of application areas, few examples of Neural Architecture Search algorithms have been inspired by methodologies, successes, and challenges in the evolution of development of embodied robots and animals. In this work, we propose to: (1) Systematically compare the quality of different neural network architectures by analyzing the amount of “embodied intelligence” in their network topologies. (2) Highlight a shortcoming of current “weight-sharing” approaches to NAS and demonstrate how an embodied perspective to brain-body co-optimization may improve search for high quality neural architectures. (3) Demonstrate how neural architectures that grow and prune their structures throughout training compare to static neural network architectures. (4) Create infrastructure to more easily integrate collaborations on real-world problems datasets into the teaching of machine learning and data science at the University of Vermont. (5) Create scientific communication materials for engaging non-STEM students in machine learning via interactive network visualizations and generative art. This project is jointly funded by the Electrical, Communications and Cyber Systems Division (ECCS) and the Established Program to Stimulate Competitive Research (EPSCoR).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
机器学习(ML)模型(如深度神经网络)的最新进展显示出解决从健康和健康到环境科学再到国防等广泛领域问题的巨大潜力。 然而,这些方法的实际使用目前需要大量的ML训练和经验来实现,因为它们对非直观和复杂的配置设置(如深度神经网络的大小和形状)非常敏感。 自动机器学习(AutoML)的子领域旨在通过创建模型和管道来帮助减少进入障碍,这些模型和管道可以根据给定问题的需求自动进行自我配置。 在AutoML中,神经架构搜索(NAS)旨在自动找到深度神经网络的理想结构。 使用ML来寻找深度神经网络的最佳形状和形式的过程大致类似于在生物中创建形状和形式的经过充分研究的进化和发展过程,或者在进化机器人领域自动找到机器人的形状和形式。 尽管这些子领域之间存在类比,而且AutoML的工作可能会对我们推动数据革命并在各种应用领域产生实际影响的能力产生巨大影响,但很少有神经架构搜索算法的例子受到具体机器人和动物发展过程中的方法论,成功和挑战的启发。 在这项工作中,我们提出:(1)系统地比较不同的神经网络架构的质量,通过分析其网络拓扑结构中的“体现智能”的数量。 (2)强调当前NAS的“权重共享”方法的缺点,并演示如何从具体的角度来实现脑-体协同优化,以改善对高质量神经架构的搜索。 (3)演示在整个训练过程中增长和修剪其结构的神经架构与静态神经网络架构的比较。 (4)创建基础设施,以便更轻松地将现实世界问题数据集的协作集成到佛蒙特大学的机器学习和数据科学教学中。(5)通过交互式网络可视化和生成艺术创建科学交流材料,让非STEM学生参与机器学习。该项目由电气,通信和网络系统部(ECCS)和刺激竞争研究的既定计划(EPSCoR)该奖项反映了NSF的法定使命,并通过使用基金会的智力价值和更广泛的评估被认为值得支持。影响审查标准。
项目成果
期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
数据更新时间:{{ journalArticles.updateTime }}
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Nicholas Cheney其他文献
Epidemiology of <em>Connectional Silence</em> in specialist serious illness conversations
- DOI:
10.1016/j.pec.2021.10.032 - 发表时间:
2022-07-01 - 期刊:
- 影响因子:
- 作者:
Cailin J. Gramling;Brigitte N. Durieux;Laurence A. Clarfeld;Ali Javed;Jeremy E. Matt;Viktoria Manukyan;Tess Braddish;Ann Wong;Joseph Wills;Laura Hirsch;Jack Straton;Nicholas Cheney;Margaret J. Eppstein;Donna M. Rizzo;Robert Gramling - 通讯作者:
Robert Gramling
The Resume Paradox: Greater Language Differences, Smaller Pay Gaps
简历悖论:语言差异越大,薪酬差距越小
- DOI:
10.48550/arxiv.2307.08580 - 发表时间:
2023 - 期刊:
- 影响因子:0
- 作者:
J. Minot;Marc Maier;Bradford Demarest;Nicholas Cheney;C. Danforth;P. Dodds;M. Frank - 通讯作者:
M. Frank
Nature is resource, playground, and gift: What artificial intelligence reveals about human–Nature relationships
自然是资源、游乐场和礼物:人工智能揭示了人与自然的关系
- DOI:
- 发表时间:
2024 - 期刊:
- 影响因子:3.7
- 作者:
Rachelle K. Gould;Bradford Demarest;Adrian Ivakhiv;Nicholas Cheney - 通讯作者:
Nicholas Cheney
EVO-SCHIRP: Evolved Secure Swarm Communications
EVO-SCHIRP:演进的安全群体通信
- DOI:
- 发表时间:
2018 - 期刊:
- 影响因子:0
- 作者:
Shaya Wolf;Rafer Cooley;Mike Borowczak;Nicholas Cheney - 通讯作者:
Nicholas Cheney
Behavioral Patterns in a Disease Spreading Simulation
疾病传播模拟中的行为模式
- DOI:
- 发表时间:
2023 - 期刊:
- 影响因子:0
- 作者:
Ollin Langle;Scott C. Merril;E. Clark;G. Bucini;Tung;T. Shrum;C. Koliba;A. Zia;Julia M. Smith;Nicholas Cheney - 通讯作者:
Nicholas Cheney
Nicholas Cheney的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Nicholas Cheney', 18)}}的其他基金
RI: Small: Collaborative Research: Evolutionary Approach to Optimal Morphology of Transformable Soft Robots
RI:小型:协作研究:可变形软机器人最佳形态的进化方法
- 批准号:
2008413 - 财政年份:2020
- 资助金额:
$ 54.99万 - 项目类别:
Standard Grant
相似海外基金
Embodied Intelligence of Music: Feedback system to enhance the musical emotion
音乐的体现智能:增强音乐情感的反馈系统
- 批准号:
22KK0157 - 财政年份:2022
- 资助金额:
$ 54.99万 - 项目类别:
Fund for the Promotion of Joint International Research (Fostering Joint International Research (B))
IRES Track I: Computational Co-Design of Physical Systems with Embodied Intelligence by Integrating Data, Simulation, and User Interface
IRES Track I:通过集成数据、仿真和用户界面,实现具有体现智能的物理系统的计算协同设计
- 批准号:
2153560 - 财政年份:2022
- 资助金额:
$ 54.99万 - 项目类别:
Standard Grant
Adaptive Mechanical Systems for Physical Embodied Intelligence Robotics
用于物理体现智能机器人的自适应机械系统
- 批准号:
RGPIN-2022-04488 - 财政年份:2022
- 资助金额:
$ 54.99万 - 项目类别:
Discovery Grants Program - Individual
From Sensing to Collaboration: Engineering, Exploring and Exploiting the Building Blocks of Embodied Intelligence - An EPSRC Programme Grant
从感知到协作:工程、探索和利用体现智能的构建模块 - EPSRC 计划资助
- 批准号:
EP/V000748/1 - 财政年份:2021
- 资助金额:
$ 54.99万 - 项目类别:
Research Grant
NRI: Liquid-Solid Metal for Embodied Intelligence in Semi-Soft, Human-Collaborative Robots
NRI:用于半软人类协作机器人中体现智能的液固金属
- 批准号:
2133027 - 财政年份:2021
- 资助金额:
$ 54.99万 - 项目类别:
Standard Grant
International Workshop on Embodied Intelligence
国际具身智能研讨会
- 批准号:
EP/T033142/1 - 财政年份:2021
- 资助金额:
$ 54.99万 - 项目类别:
Research Grant
Embodied AI and Distributed Intelligence: Being Human in the Age of Accelerated Innovation
实体人工智能和分布式智能:加速创新时代的人类
- 批准号:
2066253 - 财政年份:2018
- 资助金额:
$ 54.99万 - 项目类别:
Studentship
NRI: FND: Human-Robot Collaboration with Distributed and Embodied Intelligence
NRI:FND:具有分布式和体现智能的人机协作
- 批准号:
1734456 - 财政年份:2017
- 资助金额:
$ 54.99万 - 项目类别:
Standard Grant
Workshop on the Dynamic Interaction of Embodied Human and Machine Intelligence; Marconi State Historic Park, Marshall, California; June 2018
人类与机器智能的动态交互研讨会;
- 批准号:
1744637 - 财政年份:2017
- 资助金额:
$ 54.99万 - 项目类别:
Standard Grant
A study of a collective intelligence based museum guide system corresponding to user's embodied interaction
基于集体智慧的对应用户具身交互的博物馆导览系统研究
- 批准号:
15K01151 - 财政年份:2015
- 资助金额:
$ 54.99万 - 项目类别:
Grant-in-Aid for Scientific Research (C)