CAREER: End-to-end Constrained Optimization Learning

职业:端到端约束优化学习

基本信息

  • 批准号:
    2143706
  • 负责人:
  • 金额:
    $ 51.54万
  • 依托单位:
  • 依托单位国家:
    美国
  • 项目类别:
    Continuing Grant
  • 财政年份:
    2022
  • 资助国家:
    美国
  • 起止时间:
    2022-03-15 至 2024-02-29
  • 项目状态:
    已结题

项目摘要

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).Constrained optimization is used daily in our society with applications ranging from supply chains and logistics to electricity grids, organ exchanges, marketing campaigns, and manufacturing. Although these problems are often computationally challenging even for medium-sized instances, they constitute fundamental building blocks for the optimization of many industrial processes with profound effects on our society and economy. Yet the complexity of many constrained optimization problems often prevents them from being effectively adopted in contexts where many instances must be solved over a long-term horizon or when solutions must be produced under stringent time constraints. This project proposes a new paradigm that tightly integrates fundamental optimization techniques with machine learning algorithms to solve constraint optimization problems in real-time. This research holds the promise to create a new and transformative generation of optimization tools that solve hard constraint optimization problems under stringent time constraints leading to significant economic and societal benefits. From a scientific standpoint, this project will develop a new integration of optimization and machine learning tools that deliver high-quality solutions to large-scale hard constraint optimization problems at unprecedented computational speeds. The proposed end-to-end Constraint Optimization Learning (e2e-COL) contributes to new scientific knowledge along three main directions: (1) It accommodates the presence of domain knowledge or complex problem constraints by combining fundamental methodologies from optimization into the training cycle of deep neural networks. (2) It addresses the need of generating large datasets to train high-quality models by devising efficient data generation procedures, linking methodologies from optimization with the model learning ability, and developing semi-supervised models requiring small amounts of labeled data. (3) Finally, to scale to large problem instances, this proposal enables e2e-COL to learn decompositions and approximations of the problem structure.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.
该奖项全部或部分由2021年美国救援计划法案(公法117-2)资助。约束优化在我们的社会中每天都在使用,应用范围从供应链和物流到电网,器官交换,营销活动和制造。尽管这些问题即使对于中等规模的实例也经常具有计算挑战性,但它们构成了许多工业过程优化的基本构建块,对我们的社会和经济产生了深远的影响。然而,许多约束优化问题的复杂性往往使它们无法有效地应用于必须在长期范围内解决许多问题或必须在严格的时间限制下产生解决方案的情况。该项目提出了一种新的范式,将基本优化技术与机器学习算法紧密结合,以实时解决约束优化问题。这项研究有望创造一个新的和变革性的一代优化工具,解决严格的时间限制下的硬约束优化问题,导致显着的经济和社会效益。从科学的角度来看,该项目将开发优化和机器学习工具的新集成,以前所未有的计算速度为大规模硬约束优化问题提供高质量的解决方案。所提出的端到端约束优化学习(e2 e-COL)沿着沿着三个主要方向贡献新的科学知识:(1)它通过将优化的基本方法结合到深度神经网络的训练周期中来适应领域知识或复杂问题约束的存在。(2)它通过设计高效的数据生成程序,将优化方法与模型学习能力联系起来,以及开发需要少量标记数据的半监督模型,来满足生成大型数据集以训练高质量模型的需求。(3)最后,为了扩展到大的问题实例,该提案使e2 e-COL能够学习问题结构的分解和近似。该奖项反映了NSF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(14)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Fairness Increases Adversarial Vulnerability
  • DOI:
    10.48550/arxiv.2211.11835
  • 发表时间:
    2022-11
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Cuong Tran;Keyu Zhu;Ferdinando Fioretto;P. V. Hentenryck
  • 通讯作者:
    Cuong Tran;Keyu Zhu;Ferdinando Fioretto;P. V. Hentenryck
End-to-End Optimization and Learning for Multiagent Ensembles
  • DOI:
    10.48550/arxiv.2211.00251
  • 发表时间:
    2023
  • 期刊:
  • 影响因子:
    0
  • 作者:
    James Kotary;Vincenzo Di Vito;Ferdinando Fioretto
  • 通讯作者:
    James Kotary;Vincenzo Di Vito;Ferdinando Fioretto
Deadwooding: Robust Global Pruning for Deep Neural Networks
  • DOI:
  • 发表时间:
    2022-02
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Sawinder Kaur;Ferdinando Fioretto;Asif Salekin
  • 通讯作者:
    Sawinder Kaur;Ferdinando Fioretto;Asif Salekin
Differentiable Model Selection for Ensemble Learning
  • DOI:
    10.24963/ijcai.2023/217
  • 发表时间:
    2022-11
  • 期刊:
  • 影响因子:
    0
  • 作者:
    James Kotary;Vincenzo Di Vito;Ferdinando Fioretto
  • 通讯作者:
    James Kotary;Vincenzo Di Vito;Ferdinando Fioretto
Pruning has a disparate impact on model accuracy
剪枝对模型精度有不同的影响
{{ 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 }}

Ferdinando Fioretto其他文献

Solving DCOPs with Distributed Large Neighborhood Search
通过分布式大邻域搜索解决 DCOP
  • DOI:
  • 发表时间:
    2017
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Ferdinando Fioretto;A. Dovier;Enrico Pontelli;W. Yeoh;R. Zivan
  • 通讯作者:
    R. Zivan
Constrained-Based Differential Privacy: Releasing Optimal Power Flow Benchmarks Privately - Releasing Optimal Power Flow Benchmarks Privately
基于约束的差分隐私:私下发布最优潮流基准 - 私下发布最优潮流基准
  • DOI:
    10.1007/978-3-319-93031-2_15
  • 发表时间:
    2018
  • 期刊:
  • 影响因子:
    6.6
  • 作者:
    Ferdinando Fioretto;Pascal Van Hentenryck
  • 通讯作者:
    Pascal Van Hentenryck
Personalized Privacy Auditing and Optimization at Test Time
测试时的个性化隐私审核和优化
  • DOI:
    10.48550/arxiv.2302.00077
  • 发表时间:
    2023
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Cuong Tran;Ferdinando Fioretto
  • 通讯作者:
    Ferdinando Fioretto
A Large Neighboring Search Schema for Multi-agent Optimization
用于多智能体优化的大型邻近搜索模式
Proactive Dynamic DCOPs
主动动态 DCOP
  • DOI:
  • 发表时间:
    2016
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Khoi Hoang;Ferdinando Fioretto;Ping Hou;Makoto Yokoo;William Yeoh;Roie Zivan
  • 通讯作者:
    Roie Zivan

Ferdinando Fioretto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('Ferdinando Fioretto', 18)}}的其他基金

Collaborative Research: RI: Small: Deep Constrained Learning for Power Systems
合作研究:RI:小型:电力系统的深度约束学习
  • 批准号:
    2345528
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
Collaborative Research: RI: Small: End-to-end Learning of Fair and Explainable Schedules for Court Systems
合作研究:RI:小型:法院系统公平且可解释的时间表的端到端学习
  • 批准号:
    2232054
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
Travel: Doctoral Consortium at the 22nd International Conference on Autonomous Agents and Multiagent Systems
旅行:博士联盟出席第 22 届自主代理和多代理系统国际会议
  • 批准号:
    2246464
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
Collaborative Research: SaTC: CORE: Small: Privacy and Fairness in Critical Decision Making
协作研究:SaTC:核心:小型:关键决策中的隐私和公平
  • 批准号:
    2345483
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
Collaborative Research: Physics Informed Real-time Optimal Power Flow
合作研究:基于物理的实时最佳潮流
  • 批准号:
    2334448
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
Travel: Doctoral Consortium at the 22nd International Conference on Autonomous Agents and Multiagent Systems
旅行:博士联盟出席第 22 届自主代理和多代理系统国际会议
  • 批准号:
    2334707
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
CAREER: End-to-end Constrained Optimization Learning
职业:端到端约束优化学习
  • 批准号:
    2401285
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
Collaborative Research: Physics Informed Real-time Optimal Power Flow
合作研究:基于物理的实时最佳潮流
  • 批准号:
    2242931
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
Collaborative Research: RI: Small: End-to-end Learning of Fair and Explainable Schedules for Court Systems
合作研究:RI:小型:法院系统公平且可解释的时间表的端到端学习
  • 批准号:
    2334936
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
Collaborative Research: SaTC: CORE: Small: Privacy and Fairness in Critical Decision Making
协作研究:SaTC:核心:小型:关键决策中的隐私和公平
  • 批准号:
    2133169
  • 财政年份:
    2021
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant

相似国自然基金

真菌特异的内吞作用相关蛋白End3发挥作用的结构研究
  • 批准号:
    32000859
  • 批准年份:
    2020
  • 资助金额:
    24.0 万元
  • 项目类别:
    青年科学基金项目
从PBMC-β-END-μ-阿片受体途径探讨华蟾素治疗癌痛的外周机制
  • 批准号:
    81173612
  • 批准年份:
    2011
  • 资助金额:
    58.0 万元
  • 项目类别:
    面上项目
研究EB1(End-Binding protein 1)的癌基因特性及作用机制
  • 批准号:
    30672361
  • 批准年份:
    2006
  • 资助金额:
    24.0 万元
  • 项目类别:
    面上项目

相似海外基金

CAREER: Integrated and end-to-end machine learning pipeline for edge-enabled IoT systems: a resource-aware and QoS-aware perspective
职业:边缘物联网系统的集成端到端机器学习管道:资源感知和 QoS 感知的视角
  • 批准号:
    2340075
  • 财政年份:
    2024
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
CAREER: Radio Frequency Piezoelectric Acoustic Microsystems for Efficient and Adaptive Front-End Signal Processing
职业:用于高效和自适应前端信号处理的射频压电声学微系统
  • 批准号:
    2339731
  • 财政年份:
    2024
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
CAREER: From Dirty Data to Fair Prediction: Data Preparation Framework for End-to-End Equitable Machine Learning
职业:从脏数据到公平预测:端到端公平机器学习的数据准备框架
  • 批准号:
    2341055
  • 财政年份:
    2024
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
CAREER: End-to-End Active Region-based Heliospheric Forecasting System Using Multi-spacecraft Data and Machine Learning
职业:使用多航天器数据和机器学习的基于端对端活动区域的日光层预报系统
  • 批准号:
    2240022
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
CAREER: End-to-End Encryption for Managed Networks
职业:托管网络的端到端加密
  • 批准号:
    2236784
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
Track 1: Conference - Two Means to an End, Broadening participation research and practice - CAREER panel and workshop for DEIA leaders
轨道 1:会议 - 两种手段达到目的,扩大参与研究和实践 - DEIA 领导人的职业小组和研讨会
  • 批准号:
    2309255
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Standard Grant
CAREER: End-to-end Constrained Optimization Learning
职业:端到端约束优化学习
  • 批准号:
    2401285
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
CAREER: The end of winter: How changing snow conditions affect soil redox and biogeochemistry
职业:冬季结束:变化的雪况如何影响土壤氧化还原和生物地球化学
  • 批准号:
    2237128
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
CAREER: An Automated End-to-end Machine Learning System
职业:自动化的端到端机器学习系统
  • 批准号:
    2239351
  • 财政年份:
    2023
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
CAREER: End-User Robot Programming by Multimodal Instruction
职业:通过多模式指令进行最终用户机器人编程
  • 批准号:
    2143704
  • 财政年份:
    2022
  • 资助金额:
    $ 51.54万
  • 项目类别:
    Continuing Grant
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了