Collaborative Research: CCSS: Learning to Optimize: From New Algorithms to New Theory
合作研究:CCSS:学习优化:从新算法到新理论
基本信息
- 批准号:2113860
- 负责人:
- 金额:$ 22万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Standard Grant
- 财政年份:2021
- 资助国家:美国
- 起止时间:2021-08-15 至 2024-07-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
Solving machine learning (ML) problems requires efficient and scalable optimization algorithms. State-of-the-art general purpose algorithms often need to compute a large number of iterations and hence have limited applicability to real-time applications. To circumvent this shortcoming, learning to optimize (L2O) methods aim to learn a shorter (i.e., faster) optimization path over a task distribution at meta-training, based on the tasks’ common structures and a more global view of their geometries, and then apply the learned optimizer to new similar tasks at meta-testing. Despite extensive empirical success, the existing L2O methods perform well mainly on optimization tasks with similar structures, but likely perform poorly on out-of-distribution tasks. Furthermore, there has been little theory understanding the convergence and generalization of L2O algorithms. Thus, the proposed program will design novel L2O approaches, so that the trained optimizer can generalize to a broad range of practical tasks, particularly out-of-distribution tasks, and will have guaranteed convergence and generalization performance in L2O training and testing.Specifically, the proposed program will design new L2O approaches with both generalizability to out-of-distribution tasks and safeguarded feature for guaranteed worst-case convergence, will develop a theoretical framework for analyzing the convergence rate for L2O meta-training, and will provide comprehensive characterization of the generalization performance for L2O meta-testing. The new algorithms and theory will be evaluated over applications of on-device model adaptation in internet-of-things (IoT) systems, sparse recovery for images and wireless signals, and algorithmic adaptation in reconfiguration of communication systems. The project is anticipated to significantly mature the field of L2O, and provide training opportunities for a diverse group of students at the new intersection of optimization, machine learning, signal processing, and data science.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.
解决机器学习(ML)问题需要有效且可扩展的优化算法。最新的通用算法通常需要计算大量迭代,因此对实时应用程序的可应用性有限。为了避免这一缺点,学习优化(L2O)方法旨在根据任务的共同结构和对几何形状的更全局的看法,学习较短的(即更快)优化路径,然后将学习的优化器应用于Meta-testing的新型任务。尽管经验取得了广泛的成功,但现有的L2O方法主要在具有相似结构的优化任务上表现良好,但在分发任务上可能表现较差。此外,几乎没有理论理解L2O算法的融合和概括。拟议的计划将设计新颖的L2O方法,以便训练有素的优化器可以推广到各种各样的实际任务,尤其是分布式任务,并保证在L2O培训和测试中保证收敛性和概括性性能。具体而言,提议的计划将设计新的L2O方法,以确保范围内的AREDE SAFERISE AR DEARSTARE AR DEATERID SAFERED SAFERED SAFEREDS TEATER SAFERTIBERS TEATER SAFERD SEAFERDS TEATERS TEATERS WEATS TEATERING SERVARTS WESTIST SERVANTS WESSITITS造就,使得范围内的特征,分析L2O元训练的收敛速率以及新算法和理论的理论框架将在启用(IoT)系统中的设备模型适应(IoT)系统的应用中进行评估,图像和无线信号的稀疏恢复以及算法适应通信系统的算法适应。预计该项目将在L2O领域中显着成熟,并在新的优化,机器学习,信号处理和数据科学的交集中为潜水员组提供培训机会。该奖项反映了NSF的法定任务,并被认为是通过基金会的知识分子和更广泛的影响审查标准来通过评估来通过评估来获得的支持。
项目成果
期刊论文数量(6)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
M-L2O: Towards Generalizable Learning-to-Optimize by Test-Time Fast Self-Adaptation
- DOI:10.48550/arxiv.2303.00039
- 发表时间:2023-02
- 期刊:
- 影响因子:0
- 作者:Junjie Yang;Xuxi Chen;Tianlong Chen;Zhangyang Wang;Yitao Liang
- 通讯作者:Junjie Yang;Xuxi Chen;Tianlong Chen;Zhangyang Wang;Yitao Liang
Provably Faster Algorithms for Bilevel Optimization
- DOI:
- 发表时间:2021-06
- 期刊:
- 影响因子:0
- 作者:Junjie Yang;Kaiyi Ji;Yingbin Liang
- 通讯作者:Junjie Yang;Kaiyi Ji;Yingbin Liang
Provably Efficient Algorithm for Nonstationary Low-Rank MDPs
- DOI:10.48550/arxiv.2308.05471
- 发表时间:2023-08
- 期刊:
- 影响因子:0
- 作者:Yuan Cheng;J. Yang;Yitao Liang
- 通讯作者:Yuan Cheng;J. Yang;Yitao Liang
Theory on Forgetting and Generalization of Continual Learning
- DOI:10.48550/arxiv.2302.05836
- 发表时间:2023-02
- 期刊:
- 影响因子:0
- 作者:Sen Lin;Peizhong Ju;Yitao Liang;N. Shroff
- 通讯作者:Sen Lin;Peizhong Ju;Yitao Liang;N. Shroff
Learning to generalize provably in learning to optimize
在学习优化中学习可证明泛化
- DOI:
- 发表时间:2023
- 期刊:
- 影响因子:0
- 作者:Yang, Junjie;Chen, Tianlong;Zhu, Mingkang;He, Fengxiang;Tao, Dacheng;Liang, Yingbin;Wang, Zhangyang.
- 通讯作者:Wang, Zhangyang.
{{
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 }}
Yingbin Liang其他文献
Capacity bounds for a class of cognitive interference channels with state
一类具有状态的认知干扰信道的容量界限
- DOI:
10.1109/allerton.2011.6120223 - 发表时间:
2011 - 期刊:
- 影响因子:0
- 作者:
Ruchen Duan;Yingbin Liang - 通讯作者:
Yingbin Liang
On the Equivalence of Two Achievable Regions for the Broadcast Channel
广播频道两个可达到区域的等效性
- DOI:
10.1109/tit.2010.2090236 - 发表时间:
2011 - 期刊:
- 影响因子:2.5
- 作者:
Yingbin Liang;G. Kramer;H. Poor - 通讯作者:
H. Poor
A New Perspective of Proximal Gradient Algorithms
近端梯度算法的新视角
- DOI:
- 发表时间:
2015 - 期刊:
- 影响因子:0
- 作者:
Yi Zhou;Yingbin Liang;Lixin Shen - 通讯作者:
Lixin Shen
State-Dependent Gaussian Interference Channels: Can State Be Fully Canceled?
状态相关的高斯干扰通道:状态可以完全取消吗?
- DOI:
10.1109/tit.2016.2530665 - 发表时间:
2016 - 期刊:
- 影响因子:2.5
- 作者:
Ruchen Duan;Yingbin Liang;S. Shamai - 通讯作者:
S. Shamai
A note on inexact gradient and Hessian conditions for cubic regularized Newton's method
关于三次正则牛顿法的不精确梯度和 Hessian 条件的说明
- DOI:
10.1016/j.orl.2019.01.009 - 发表时间:
2019 - 期刊:
- 影响因子:0
- 作者:
Zhe Wang;Yi Zhou;Yingbin Liang;Guanghui Lan - 通讯作者:
Guanghui Lan
Yingbin Liang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('Yingbin Liang', 18)}}的其他基金
RINGS: A Deep Reinforcement Learning Enabled Large-scale UAV Network with Distributed Navigation, Mobility Control, and Resilience
RINGS:深度强化学习支持的大规模无人机网络,具有分布式导航、移动控制和弹性
- 批准号:
2148253 - 财政年份:2022
- 资助金额:
$ 22万 - 项目类别:
Continuing Grant
Collaborative Research: SCALE MoDL: Adaptivity of Deep Neural Networks
合作研究:SCALE MoDL:深度神经网络的适应性
- 批准号:
2134145 - 财政年份:2021
- 资助金额:
$ 22万 - 项目类别:
Continuing Grant
CIF: Small: Collaborative Research: Acceleration Algorithms for Large-scale Nonconvex Optimization
CIF:小型:协作研究:大规模非凸优化的加速算法
- 批准号:
1909291 - 财政年份:2019
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
CIF: Medium: Collaborative Research: Theory of Optimization Geometry and Algorithms for Neural Networks
CIF:媒介:协作研究:神经网络优化几何理论和算法
- 批准号:
1900145 - 财政年份:2019
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
CIF: Small: Collaborative Research: Network Event Detection with Multistream Observations
CIF:小型:协作研究:通过多流观察进行网络事件检测
- 批准号:
1801855 - 财政年份:2017
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
CIF: Medium: Collaborative Research: Nonconvex Optimization for High-Dimensional Signal Estimation: Theory and Fast Algorithms
CIF:中:协作研究:高维信号估计的非凸优化:理论和快速算法
- 批准号:
1761506 - 财政年份:2017
- 资助金额:
$ 22万 - 项目类别:
Continuing Grant
CIF: Medium: Collaborative Research: Nonconvex Optimization for High-Dimensional Signal Estimation: Theory and Fast Algorithms
CIF:中:协作研究:高维信号估计的非凸优化:理论和快速算法
- 批准号:
1704169 - 财政年份:2017
- 资助金额:
$ 22万 - 项目类别:
Continuing Grant
Management of Mobile Phone Sensing via Sparse Learning
通过稀疏学习管理手机传感
- 批准号:
1818904 - 财政年份:2017
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
CIF: Small: Collaborative Research: Secret Key Generation Under Resource Constraints
CIF:小型:协作研究:资源限制下的密钥生成
- 批准号:
1801846 - 财政年份:2017
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
CIF: Small: Collaborative Research: Secret Key Generation Under Resource Constraints
CIF:小型:协作研究:资源限制下的密钥生成
- 批准号:
1618127 - 财政年份:2016
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
相似国自然基金
钛基骨植入物表面电沉积镁氢涂层及其促成骨性能研究
- 批准号:52371195
- 批准年份:2023
- 资助金额:50 万元
- 项目类别:面上项目
CLMP介导Connexin45-β-catenin复合体对先天性短肠综合征的致病机制研究
- 批准号:82370525
- 批准年份:2023
- 资助金额:49 万元
- 项目类别:面上项目
人工局域表面等离激元高灵敏传感及其系统小型化的关键技术研究
- 批准号:62371132
- 批准年份:2023
- 资助金额:49 万元
- 项目类别:面上项目
优先流对中俄原油管道沿线多年冻土水热稳定性的影响机制研究
- 批准号:42301138
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
用于稳定锌负极的界面层/电解液双向调控研究
- 批准号:52302289
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
相似海外基金
Collaborative Research: ECCS-CCSS Core: Resonant-Beam based Optical-Wireless Communication
合作研究:ECCS-CCSS核心:基于谐振光束的光无线通信
- 批准号:
2332172 - 财政年份:2024
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
Collaborative Research: ECCS-CCSS Core: Resonant-Beam based Optical-Wireless Communication
合作研究:ECCS-CCSS核心:基于谐振光束的光无线通信
- 批准号:
2332173 - 财政年份:2024
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
Collaborative Research: CCSS: Continuous Facial Sensing and 3D Reconstruction via Single-ear Wearable Biosensors
合作研究:CCSS:通过单耳可穿戴生物传感器进行连续面部传感和 3D 重建
- 批准号:
2401415 - 财政年份:2023
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
Collaborative Research: CCSS: When RFID Meets AI for Occluded Body Skeletal Posture Capture in Smart Healthcare
合作研究:CCSS:当 RFID 与人工智能相遇,用于智能医疗保健中闭塞的身体骨骼姿势捕获
- 批准号:
2245607 - 财政年份:2023
- 资助金额:
$ 22万 - 项目类别:
Standard Grant
Collaborative Research: CCSS: Hierarchical Federated Learning over Highly-Dense and Overlapping NextG Wireless Deployments: Orchestrating Resources for Performance
协作研究:CCSS:高密度和重叠的 NextG 无线部署的分层联合学习:编排资源以提高性能
- 批准号:
2319780 - 财政年份:2023
- 资助金额:
$ 22万 - 项目类别:
Standard Grant