FRG: Collaborative Research: Robust, Efficient, and Private Deep Learning Algorithms
FRG:协作研究:稳健、高效、私密的深度学习算法
基本信息
- 批准号:1952339
- 负责人:
- 金额:$ 43.48万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Standard Grant
- 财政年份:2020
- 资助国家:美国
- 起止时间:2020-08-15 至 2024-07-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
This project develops robust, accurate, and efficient next-generation deep learning algorithms with data privacy and theoretical guarantees for solving challenging artificial intelligence (AI) problems. The methods will have robustness to adversarial attacks with theoretical guarantees. The project will push artificial intelligence gains in performance and privacy to mobile devices. A broad range of applications includes autonomous driving, drug and material discovery, medical treatment planning, national defense, privacy-preserving machine learning at the edge, federated learning, and also blockchain. Moreover, the developed tools will significantly scale the existing scientific simulations to ultra-large scale and high-dimensional scenarios. This project will partially support one graduate student per year at each campus.Our approach toward trustworthy deep learning is theoretically principled by modern partial differential equations and optimization algorithms and theories. The project involves new algorithmic and theoretical techniques to tackle graph representation in high-dimensional non-convex, non-smooth AI settings. In particular, the project will study (1) developing adversarial robust deep learning algorithms and their theoretical foundations; (2) improving the accuracy of deep learning leveraging new stochastic optimization and principled neural network unit design assisted neural architecture search; (3) advancing deep neural networks compression with algorithms and hardware co-design; (4) designing new data privacy mechanisms to optimally tradeoff between utility and privacy; (5) inventing new quantitative analysis tools to decipher the mysteries of deep learning theoretical challenges; (6) quantifying uncertainties of sophisticated deep learning algorithms. The project trains a diverse body of graduate and undergraduate students at UC Irvine, UCLA, and University of Utah through collaborative education and research activities in applied mathematics, computer science, data science, and general biological, physical, and sociological disciplines.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.
该项目开发强大,准确,高效的下一代深度学习算法,具有数据隐私和理论保证,以解决具有挑战性的人工智能(AI)问题。该方法对对抗性攻击具有鲁棒性,并有理论保证。该项目将把人工智能在性能和隐私方面的进步推向移动设备。广泛的应用包括自动驾驶、药物和材料发现、医疗计划、国防、边缘保护隐私的机器学习、联邦学习以及区块链。此外,开发的工具将显着扩展现有的科学模拟到超大规模和高维场景。该项目将在每个校区每年资助一名研究生。我们对可信赖深度学习的方法在理论上是以现代偏微分方程和优化算法和理论为原则的。该项目涉及新的算法和理论技术,以解决高维非凸、非光滑人工智能设置中的图形表示。特别是,该项目将研究(1)开发对抗性鲁棒深度学习算法及其理论基础;(2)利用新的随机优化和原则神经网络单元设计辅助神经结构搜索来提高深度学习的准确性;(3)通过算法和硬件协同设计推进深度神经网络压缩;(4)设计新的数据隐私机制,在效用和隐私之间进行最佳权衡;(5)发明新的定量分析工具,破解深度学习理论挑战的奥秘;(6)量化复杂深度学习算法的不确定性。该项目通过应用数学、计算机科学、数据科学以及一般生物学、物理学和社会学学科的合作教育和研究活动,在加州大学欧文分校、加州大学洛杉矶分校和犹他大学培养了一批多样化的研究生和本科生。该奖项反映了美国国家科学基金会的法定使命,并通过使用基金会的知识价值和更广泛的影响审查标准进行评估,被认为值得支持。
项目成果
期刊论文数量(21)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Graph-based active learning for semi-supervised classification of SAR data
- DOI:10.1117/12.2618847
- 发表时间:2022-03
- 期刊:
- 影响因子:0
- 作者:Kevin Miller;John Mauro;Jason Setiadi;Xoaquin Baca;Zhan Shi;J. Calder;A. Bertozzi
- 通讯作者:Kevin Miller;John Mauro;Jason Setiadi;Xoaquin Baca;Zhan Shi;J. Calder;A. Bertozzi
MomentumRNN: Integrating Momentum into Recurrent Neural Networks
- DOI:
- 发表时间:2020-06
- 期刊:
- 影响因子:0
- 作者:T. Nguyen;Richard Baraniuk;A. Bertozzi;S. Osher;Baorui Wang
- 通讯作者:T. Nguyen;Richard Baraniuk;A. Bertozzi;S. Osher;Baorui Wang
Privacy-Preserving Federated Multi-Task Linear Regression: A One-Shot Linear Mixing Approach Inspired By Graph Regularization
- DOI:10.1109/icassp43922.2022.9746007
- 发表时间:2022-05
- 期刊:
- 影响因子:0
- 作者:Harlin Lee;A. Bertozzi;J. Kovacevic;Yuejie Chi
- 通讯作者:Harlin Lee;A. Bertozzi;J. Kovacevic;Yuejie Chi
Scheduled Restart Momentum for Accelerated Stochastic Gradient Descent
- DOI:10.1137/21m1453311
- 发表时间:2020-02
- 期刊:
- 影响因子:0
- 作者:Bao Wang;T. Nguyen;A. Bertozzi;Richard Baraniuk;S. Osher
- 通讯作者:Bao Wang;T. Nguyen;A. Bertozzi;Richard Baraniuk;S. Osher
Graph interpolating activation improves both natural and robust accuracies in data-efficient deep learning
- DOI:10.1017/s0956792520000406
- 发表时间:2019-07
- 期刊:
- 影响因子:1.9
- 作者:Bao Wang;S. Osher
- 通讯作者:Bao Wang;S. Osher
{{
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 }}
Andrea Bertozzi其他文献
Incorporating Texture Features into Optical Flow for Atmospheric Wind Velocity Estimation
将纹理特征纳入光流中进行大气风速估计
- DOI:
- 发表时间:
2023 - 期刊:
- 影响因子:0
- 作者:
Joel Barnett;Andrea Bertozzi;L. Vese;Igor Yanovsky - 通讯作者:
Igor Yanovsky
Encased Cantilevers and Alternative Scan Algorithms for Ultra-Gantle High Speed Atomic Force Microscopy
- DOI:
10.1016/j.bpj.2011.11.3193 - 发表时间:
2012-01-31 - 期刊:
- 影响因子:
- 作者:
Paul Ashby;Dominik Ziegler;Andreas Frank;Sindy Frank;Alex Chen;Travis Meyer;Rodrigo Farnham;Nen Huynh;Ivo Rangelow;Jen-Mei Chang;Andrea Bertozzi - 通讯作者:
Andrea Bertozzi
Andrea Bertozzi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('Andrea Bertozzi', 18)}}的其他基金
Collaborative Research: RAPID: Rapid computational modeling of wildfires and management with emphasis on human activity
合作研究:RAPID:野火和管理的快速计算建模,重点关注人类活动
- 批准号:
2345256 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
ATD: Active Learning Activity Detection in Multiplex Networks of Geospatial-Cyber-Temporal Data
ATD:地理空间网络时空数据多重网络中的主动学习活动检测
- 批准号:
2318817 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
Collaborative Research: Differential Equations Motivated Multi-Agent Sequential Deep Learning: Algorithms, Theory, and Validation
协作研究:微分方程驱动的多智能体序列深度学习:算法、理论和验证
- 批准号:
2152717 - 财政年份:2022
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
RAPID: Analysis of Multiscale Network Models for the Spread of COVID-19
RAPID:针对 COVID-19 传播的多尺度网络模型分析
- 批准号:
2027438 - 财政年份:2020
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
ATD: Algorithms for Threat Detection in Knowledge Graphs
ATD:知识图中的威胁检测算法
- 批准号:
2027277 - 财政年份:2020
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
NRT-HDR: Modeling and Understanding Human Behavior: Harnessing Data from Genes to Social Networks
NRT-HDR:建模和理解人类行为:利用从基因到社交网络的数据
- 批准号:
1829071 - 财政年份:2018
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
ATD: Sparsity Models for Forecasting Spatio-Temporal Human Dynamics
ATD:预测时空人类动力学的稀疏模型
- 批准号:
1737770 - 财政年份:2017
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
Extreme-scale algorithms for geometric graphical data models in imaging, social and network science
成像、社会和网络科学中几何图形数据模型的超大规模算法
- 批准号:
1417674 - 财政年份:2014
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant
Collaborative Research: Modeling, Analysis, and Control of the Spatio-temporal Dynamics of Swarm Robotic Systems
协作研究:群体机器人系统时空动力学的建模、分析和控制
- 批准号:
1435709 - 财政年份:2014
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
Particle laden flows - theory, analysis and experiment
颗粒负载流 - 理论、分析和实验
- 批准号:
1312543 - 财政年份:2013
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant
相似海外基金
FRG: Collaborative Research: New birational invariants
FRG:协作研究:新的双有理不变量
- 批准号:
2244978 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant
FRG: Collaborative Research: Singularities in Incompressible Flows: Computer Assisted Proofs and Physics-Informed Neural Networks
FRG:协作研究:不可压缩流中的奇异性:计算机辅助证明和物理信息神经网络
- 批准号:
2245017 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
FRG: Collaborative Research: Variationally Stable Neural Networks for Simulation, Learning, and Experimental Design of Complex Physical Systems
FRG:协作研究:用于复杂物理系统仿真、学习和实验设计的变稳定神经网络
- 批准号:
2245111 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant
FRG: Collaborative Research: Variationally Stable Neural Networks for Simulation, Learning, and Experimental Design of Complex Physical Systems
FRG:协作研究:用于复杂物理系统仿真、学习和实验设计的变稳定神经网络
- 批准号:
2245077 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant
FRG: Collaborative Research: Singularities in Incompressible Flows: Computer Assisted Proofs and Physics-Informed Neural Networks
FRG:协作研究:不可压缩流中的奇异性:计算机辅助证明和物理信息神经网络
- 批准号:
2244879 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
FRG: Collaborative Research: New Birational Invariants
FRG:合作研究:新的双理性不变量
- 批准号:
2245171 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant
FRG: Collaborative Research: Singularities in Incompressible Flows: Computer Assisted Proofs and Physics-Informed Neural Networks
FRG:协作研究:不可压缩流中的奇异性:计算机辅助证明和物理信息神经网络
- 批准号:
2403764 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
FRG: Collaborative Research: Singularities in Incompressible Flows: Computer Assisted Proofs and Physics-Informed Neural Networks
FRG:协作研究:不可压缩流中的奇异性:计算机辅助证明和物理信息神经网络
- 批准号:
2245021 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Standard Grant
FRG: Collaborative Research: Variationally Stable Neural Networks for Simulation, Learning, and Experimental Design of Complex Physical Systems
FRG:协作研究:用于复杂物理系统仿真、学习和实验设计的变稳定神经网络
- 批准号:
2245097 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant
FRG: Collaborative Research: Variationally Stable Neural Networks for Simulation, Learning, and Experimental Design of Complex Physical Systems
FRG:协作研究:用于复杂物理系统仿真、学习和实验设计的变稳定神经网络
- 批准号:
2245147 - 财政年份:2023
- 资助金额:
$ 43.48万 - 项目类别:
Continuing Grant