CAREER: Interplay between Control Theory and Machine Learning
职业:控制理论和机器学习之间的相互作用
基本信息
- 批准号:2048168
- 负责人:
- 金额:$ 50万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Continuing Grant
- 财政年份:2021
- 资助国家:美国
- 起止时间:2021-03-01 至 2026-02-28
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
Control and machine learning are two high-impact research areas. Both are important for managing complex systems such as self-driving vehicles, humanoid robotics, smart buildings, and automated healthcare. This CAREER proposal aims at building fundamental connections between control theory and machine learning. On one hand, control theory provides mathematically rigorous tools for addressing the robustness requirement of modern safety-critical systems such as commercial aircraft and nuclear plants. On the other hand, machine learning techniques have been used to achieve the state-of-the-art performance for many artificial intelligence tasks in computer vision, natural language processing, and Go. A rapprochement of control theory and machine learning will significantly broaden the class of engineering problems that can be solved efficiently. This proposal aims at reconciling these two areas with a comprehensive interdisciplinary approach that spans and connects the forefronts of robust control theory, nonlinear system theory, jump system theory, supervised learning, reinforcement learning, imitation learning, semidefinite programming, and non-convex optimization. The proposed research will lay the theoretical foundation for reliable integration of control and machine learning in modern safety-critical intelligent systems. The research progress will promote multidisciplinary collaborations and benefit the researchers from learning, control, optimization, artificial intelligence, autonomy, and robotics. In addition, the research will be strongly coupled with educational developments which will promote students from different departments to develop solid multidisciplinary proficiency. New course materials resulting from the research will provide new concepts and ideas to inspire the next generation of academic and industrial leaders.This proposal takes an interdisciplinary perspective on machine learning and control. The proposed research is centered around two thrusts. The first thrust focuses on tailoring control theory to unify, streamline, and automate the analysis and design of machine learning algorithms. Specifically, algorithms in supervised/reinforcement/unsupervised learning will be modeled as Markovian jump systems and nonlinear systems which have been extensively studied in controls literature. The combination of this idea with modern control-theoretical tools such as stochastic dissipation inequalities will pave the way for a unified principled approach to the design of high-performance algorithmic pipelines in machine learning. The second thrust focuses on borrowing recent results in non-convex learning to push control theory beyond the convex optimization regime. The recently developed non-convex learning theory will be leveraged to derive various theoretical guarantees for non-convex optimization problems in control. The proposed research is expected to advance the state-of-the-art algorithms in large-scale control and broaden the class of nonlinear/robust control problems that can be solved with guarantees. The proposed research covers both “control for learning” and “learning for control,” deepening the connections of control and learning by showing that the techniques used by each side can be explored to impact the other side.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.
控制和机器学习是两个影响力很大的研究领域。这两者对于管理自动驾驶汽车、人形机器人、智能建筑和自动医疗等复杂系统都很重要。这个CAREER提案旨在建立控制理论和机器学习之间的基本联系。一方面,控制理论为解决现代安全关键系统(如商用飞机和核电站)的鲁棒性要求提供了数学上严格的工具。另一方面,机器学习技术已被用于实现计算机视觉,自然语言处理和Go中许多人工智能任务的最先进性能。控制理论和机器学习的融合将大大拓宽可以有效解决的工程问题的类别。该提案旨在通过一种全面的跨学科方法来协调这两个领域,该方法跨越并连接了鲁棒控制理论,非线性系统理论,跳跃系统理论,监督学习,强化学习,模仿学习,半定规划和非凸优化的前沿。该研究将为现代安全关键智能系统中控制和机器学习的可靠集成奠定理论基础。研究进展将促进多学科合作,并使研究人员从学习,控制,优化,人工智能,自主和机器人技术中受益。此外,该研究将与教育发展紧密结合,这将促进来自不同部门的学生发展坚实的多学科能力。从研究中产生的新课程材料将提供新的概念和想法,以激励下一代学术和工业领导者。该提案从跨学科的角度看待机器学习和控制。拟议的研究围绕两个重点。第一个重点是裁剪控制理论,以统一,简化和自动化机器学习算法的分析和设计。具体来说,监督/强化/无监督学习中的算法将被建模为马尔可夫跳跃系统和非线性系统,这些系统在控制文献中已经得到了广泛的研究。将这一思想与现代控制理论工具(如随机耗散不等式)相结合,将为机器学习中高性能算法管道设计的统一原则方法铺平道路。第二个重点是借用非凸学习的最新成果,将控制理论推向凸优化领域之外。近年来发展起来的非凸学习理论将被用来为控制中的非凸优化问题提供各种理论保证。所提出的研究有望推进大规模控制中的最先进算法,并拓宽可以用保证解决的非线性/鲁棒控制问题的类别。该研究计划涵盖“为学习而控制”和“为控制而学习”两个方面,通过展示双方所使用的技术可以对另一方产生影响,从而加深控制和学习之间的联系。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(12)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Derivative-Free Policy Optimization for Linear Risk-Sensitive and Robust Control Design: Implicit Regularization and Sample Complexity
线性风险敏感和鲁棒控制设计的无导数策略优化:隐式正则化和样本复杂性
- DOI:
- 发表时间:2021
- 期刊:
- 影响因子:0
- 作者:Zhang, Kaiqing;Zhang, Xiangyuan;Hu, Bin;Başar, Tamer
- 通讯作者:Başar, Tamer
Provable Acceleration of Heavy Ball beyond Quadratics for a Class of Polyak-Lojasiewicz Functions when the Non-Convexity is Averaged-Out
非凸性平均时一类 Polyak-Lojasiewicz 函数重球超越二次函数的可证明加速度
- DOI:
- 发表时间:2022
- 期刊:
- 影响因子:0
- 作者:Wang, Jun-Kun;Lin, Chi-Heng;Wibisono, Andre;Hu, Bin
- 通讯作者:Hu, Bin
Convex Programs and Lyapunov Functions for Reinforcement Learning: A Unified Perspective on the Analysis of Value-Based Methods
- DOI:10.23919/acc53348.2022.9867291
- 发表时间:2022-02
- 期刊:
- 影响因子:0
- 作者:Xing-ming Guo;B. Hu
- 通讯作者:Xing-ming Guo;B. Hu
Connectivity of the Feasible and Sublevel Sets of Dynamic Output Feedback Control With Robustness Constraints
具有鲁棒性约束的动态输出反馈控制的可行集和子级集的连通性
- DOI:10.1109/lcsys.2022.3188008
- 发表时间:2023
- 期刊:
- 影响因子:3
- 作者:Hu, Bin;Zheng, Yang
- 通讯作者:Zheng, Yang
Global Convergence of Direct Policy Search for State-Feedback $\mathcal{H}_\infty$ Robust Control: A Revisit of Nonsmooth Synthesis with Goldstein Subdifferential
状态反馈 $mathcal{H}_infty$ 鲁棒控制的直接策略搜索的全局收敛:重新审视 Goldstein 次微分的非平滑综合
- DOI:
- 发表时间:2022
- 期刊:
- 影响因子:0
- 作者:Guo, Xingang;Hu, Bin
- 通讯作者:Hu, Bin
{{
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 }}
Bin Hu其他文献
Fast Scalar Quadratic Maximum Likelihood Estimators for the CMB B-mode Power Spectrum
CMB B 模式功率谱的快速标量二次最大似然估计器
- DOI:
10.3847/1538-4365/ac18c9 - 发表时间:
2021-04 - 期刊:
- 影响因子:0
- 作者:
Jiming Chen;Shamik Ghosh;Hao Liu;Larissa Santos;Wenjuan Fang;Siyu Li;Yang Liu;Hong Li;Jiaxin Wang;Le Zhang;Bin Hu;Wen Zhao - 通讯作者:
Wen Zhao
Three copper(II) complexes of 4-formylbenzoate obtained from the degradation of Schiff-base C=N double bonds
希夫碱 C=N 双键降解得到的 4-甲酰基苯甲酸的三种铜(II)配合物
- DOI:
- 发表时间:
2012 - 期刊:
- 影响因子:0
- 作者:
Bin Hu;Xiao-Xu Wang;Jiao Geng;Wei Huang - 通讯作者:
Wei Huang
A Low-Frequency Approximation PEEC Model for Thin-Wire Grounding Systems in Half-Space
半空间细线接地系统的低频近似 PEEC 模型
- DOI:
10.1109/tpwrd.2022.3212778 - 发表时间:
2023-04 - 期刊:
- 影响因子:0
- 作者:
WeiPing Ye;Weidong Liu;Nianwen Xiang;Kejie Li;Lingyun Cheng;Bin Hu - 通讯作者:
Bin Hu
Thermal Boundary Portrait of Wind Power Converter under Low Voltage Ride-Through with Maximum Reactive Power Capability
具有最大无功功率能力的低电压穿越下风电变流器的热边界图
- DOI:
10.1109/ecce53617.2023.10362676 - 发表时间:
2023 - 期刊:
- 影响因子:0
- 作者:
Siyu Cao;Zhenyan Deng;Han Wang;Ke Ma;Yunfeng Cao;Xu Cai;Jiaoze Chen;Bin Hu;Zhenyuan Zhang - 通讯作者:
Zhenyuan Zhang
Improved performance of photomultiplication polymer photodetectors by adjusting P3HT molecular arrangement
通过调整 P3HT 分子排列提高光电倍增聚合物光电探测器的性能
- DOI:
- 发表时间:
- 期刊:
- 影响因子:9.5
- 作者:
Wenbin Wang;Fujun Zhang;Lingliang Li;Mile Gao;Bin Hu - 通讯作者:
Bin Hu
Bin Hu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('Bin Hu', 18)}}的其他基金
Structural basis of the Scc2/cohesin interaction and its implication on cohesin loading
Scc2/粘连蛋白相互作用的结构基础及其对粘连蛋白负载的影响
- 批准号:
BB/S002537/2 - 财政年份:2020
- 资助金额:
$ 50万 - 项目类别:
Research Grant
Structural basis of the Scc2/cohesin interaction and its implication on cohesin loading
Scc2/粘连蛋白相互作用的结构基础及其对粘连蛋白负载的影响
- 批准号:
BB/S002537/1 - 财政年份:2019
- 资助金额:
$ 50万 - 项目类别:
Research Grant
Exploring Spin-Orbital Coupling Effects: 3D to 2D Perovskite Solar Cells
探索自旋轨道耦合效应:3D 至 2D 钙钛矿太阳能电池
- 批准号:
1911659 - 财政年份:2019
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
Addressing Dynamic Donor:Acceptor and Electrode Interfaces in Organic Bulk-Heterojunction and Perovskite Solar Cells Under Device-Operating Condition
解决器件工作条件下有机体异质结和钙钛矿太阳能电池中的动态供体:受体和电极界面
- 批准号:
1438181 - 财政年份:2014
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
Workshop on Next-Generation High-Efficiency Organic Solar Cells: Opportunities and Challenges. To be Held on September 6-7, 2012 at a Hotel (TBD) in Arlington, Virginia.
下一代高效有机太阳能电池研讨会:机遇与挑战。
- 批准号:
1239169 - 财政年份:2012
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
Magneto-Optical Studies of Charge dissociation, Transport, and Collection in Organic Solar Cells
有机太阳能电池中电荷解离、传输和收集的磁光研究
- 批准号:
1102011 - 财政年份:2011
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
Planning Visits and Workshops in Brazil towards US-Brazil International Collaboration in Emerging Science: Magnetic Field Effects in Non-Magnetic Organic Semiconductors
计划在巴西举行访问和研讨会,以促进美巴西新兴科学领域的国际合作:非磁性有机半导体中的磁场效应
- 批准号:
0929566 - 财政年份:2009
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
CAREER: Research and Education in Development of Organic Spintronics Based on Spin Injection and Modification of Spin-Orbital Coupling in Magnetic Organic Light-Emitting Diodes
职业:基于磁性有机发光二极管中自旋注入和自旋轨道耦合修饰的有机自旋电子学发展的研究和教育
- 批准号:
0644945 - 财政年份:2007
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
SGER: Spin Injection from Ferromagnetic Nanodot Electrode to Organic Semiconducting Conjugated Polymers
SGER:从铁磁纳米点电极到有机半导体共轭聚合物的自旋注射
- 批准号:
0551914 - 财政年份:2005
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
SGER: Spin-Polarized Electronic Processes in Conjugated Polymer Optoelectronic Devices
SGER:共轭聚合物光电器件中的自旋极化电子过程
- 批准号:
0521474 - 财政年份:2005
- 资助金额:
$ 50万 - 项目类别:
Standard Grant
相似海外基金
CAREER: Interplay between Convex and Nonconvex Optimization for Control
职业:凸和非凸优化控制之间的相互作用
- 批准号:
2340713 - 财政年份:2024
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: The molecular interplay between integrin adhesion, allostery and shape-shifting
职业:整合素粘附、变构和变形之间的分子相互作用
- 批准号:
2239492 - 财政年份:2023
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
Career: Moving During Deliberation-Online Interplay Between Deciding and Acting
职业生涯:深思熟虑中的行动——决策与行动之间的在线相互作用
- 批准号:
2234748 - 财政年份:2023
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: Physics at the Interplay between the Higgs Boson and the Top Quark
职业:希格斯玻色子和顶夸克之间相互作用的物理学
- 批准号:
2046280 - 财政年份:2021
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: Structural Interplay Between Chromatin Remodeling and Lentiviral Integration
职业:染色质重塑和慢病毒整合之间的结构相互作用
- 批准号:
2048095 - 财政年份:2021
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: Elucidating the Interplay Between Exciton Dynamics and Symmetry-Breaking Charge Transfer Through Multidimensional Visible and Mid-Infrared Spectroscopies
职业:通过多维可见光和中红外光谱阐明激子动力学与对称破缺电荷转移之间的相互作用
- 批准号:
2047614 - 财政年份:2021
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: Understanding the interplay between lipid composition and biomolecule transport in biological membranes
职业:了解生物膜中脂质成分与生物分子运输之间的相互作用
- 批准号:
1942581 - 财政年份:2020
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: Integrated X-ray Ptychography and Optical Luminescence to Understand the Interplay between Local Structural Dynamics and Optoelectronic States
职业:集成 X 射线叠层成像和光学发光以了解局部结构动力学和光电态之间的相互作用
- 批准号:
1848371 - 财政年份:2019
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: The Interplay between Combinatorial Optimization and Algorithmic Convex Geometry
职业:组合优化和算法凸几何之间的相互作用
- 批准号:
1749609 - 财政年份:2018
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant
CAREER: The Mechanistic Interplay Between Biofilm Metabolism and Morphogenesis
职业:生物膜代谢与形态发生之间的机制相互作用
- 批准号:
1553023 - 财政年份:2015
- 资助金额:
$ 50万 - 项目类别:
Continuing Grant