Theoretical Guarantees of Machine Learning Methods for High Dimensional Partial Differential Equations: Numerical Analysis and Uncertainty Quantification

高维偏微分方程机器学习方法的理论保证:数值分析和不确定性量化

基本信息

  • 批准号:
    2107934
  • 负责人:
  • 金额:
    $ 20.5万
  • 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:
    2021-08-01 至 2023-09-30
  • 项目状态:
    已结题

项目摘要

Machine learning algorithms have achieved tremendous empirical successes in providing practical answers to various applications in our everyday life, such as face recognition and autonomous driving. This project will develop theoretical foundations of machine learning methods for applied problems in science and engineering. The research will play a principal role in determining predictive power and quantifying the robustness and stability of the machine learning methodology in applications. The investigator will mentor graduate and undergraduate students to work on both theoretical and applied aspects of the project. The investigator will provide outreach to high school students with an introductory course on Data Science and develop new mathematical machine learning courses at both graduate and advanced undergraduate levels.The project will develop a systematic mathematical framework for analyzing neural network-based methods for solving partial differential equations (PDEs), emphasizing their high-dimensional performance and uncertainty quantification. The investigator will work on two projects. The first is to derive new dimension-explicit convergence estimates on the generalization error and training dynamics of neural network solutions. This relies on establishing a new regularity theory for PDEs in new complexity-based function spaces tied to neural networks. The second objective is to quantify the uncertainty in the neural network prediction in a Bayesian framework. The research will focus on studying the frequentist performance and the scalable posterior computation of the Bayesian neural networks for solving high dimensional PDEs.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.
机器学习算法在为我们日常生活中的各种应用(如人脸识别和自动驾驶)提供实际答案方面取得了巨大的经验成功。该项目将为科学和工程中的应用问题开发机器学习方法的理论基础。该研究将在确定预测能力和量化应用中机器学习方法的鲁棒性和稳定性方面发挥主要作用。研究者将指导研究生和本科生在项目的理论和应用方面工作。该研究员将为高中生提供数据科学入门课程,并为研究生和高级本科生开发新的数学机器学习课程。该项目将开发一个系统的数学框架,用于分析求解偏微分方程(PDEs)的基于神经网络的方法,强调其高维性能和不确定性量化。调查员将从事两个项目。首先是在神经网络解的泛化误差和训练动力学上推导新的维显式收敛估计。这依赖于在与神经网络相关的新的基于复杂性的函数空间中为偏微分方程建立新的正则性理论。第二个目标是在贝叶斯框架下量化神经网络预测中的不确定性。本研究将重点研究贝叶斯神经网络在求解高维偏微分方程中的频率性能和可扩展后验计算。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(5)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
On the Representation of Solutions to Elliptic PDEs in Barron Spaces
  • DOI:
  • 发表时间:
    2021-06
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Ziang Chen;Jianfeng Lu;Yulong Lu
  • 通讯作者:
    Ziang Chen;Jianfeng Lu;Yulong Lu
Exponential-Wrapped Distributions on Symmetric Spaces
对称空间上的指数包裹分布
Solving multiscale steady radiative transfer equation using neural networks with uniform stability
  • DOI:
    10.1007/s40687-022-00345-z
  • 发表时间:
    2021-10
  • 期刊:
  • 影响因子:
    1.2
  • 作者:
    Yulong Lu;Li Wang;Wuzhe Xu
  • 通讯作者:
    Yulong Lu;Li Wang;Wuzhe Xu
A Regularity Theory for Static Schrödinger Equations on \(\boldsymbol{\mathbb{R}^d}\) in Spectral Barron Spaces
谱巴伦空间中(oldsymbol{mathbb{R}^d})静态薛定谔方程的正则理论
  • DOI:
    10.1137/22m1478719
  • 发表时间:
    2023
  • 期刊:
  • 影响因子:
    2
  • 作者:
    Chen, Ziang;Lu, Jianfeng;Lu, Yulong;Zhou, Shengxuan
  • 通讯作者:
    Zhou, Shengxuan
{{ 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 }}

Yulong Lu其他文献

Fully discretized Sobolev gradient flow for the Gross-Pitaevskii eigenvalue problem
Gross-Pitaevskii 特征值问题的完全离散 Sobolev 梯度流
  • DOI:
  • 发表时间:
    2024
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Ziang Chen;Jian Lu;Yulong Lu;Xiangxiong Zhang
  • 通讯作者:
    Xiangxiong Zhang
Asymptotic analysis and computations of probability measures
  • DOI:
  • 发表时间:
    2017-06
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Yulong Lu
  • 通讯作者:
    Yulong Lu
Experimental Study on the Chemical and Geo-Chemical Properties of a Fe3O4 Dominated Ore
Fe3O4为主矿石化学和地球化学性质的实验研究
Quantitative Propagation of Chaos in a Bimolecular Chemical Reaction-Diffusion Model
双分子化学反应扩散模型中混沌的定量传播
  • DOI:
    10.1137/19m1287687
  • 发表时间:
    2019
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Tau Shean Lim;Yulong Lu;J. Nolen
  • 通讯作者:
    J. Nolen
Transfer Learning Enhanced DeepONet for Long-Time Prediction of Evolution Equations
迁移学习增强 DeepONet 用于进化方程的长时间预测

Yulong Lu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('Yulong Lu', 18)}}的其他基金

Theoretical Guarantees of Machine Learning Methods for High Dimensional Partial Differential Equations: Numerical Analysis and Uncertainty Quantification
高维偏微分方程机器学习方法的理论保证:数值分析和不确定性量化
  • 批准号:
    2343135
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Standard Grant

相似海外基金

CAREER: Theoretical and Computational Advances for Enabling Robust Numerical Guarantees in Linear and Mixed Integer Programming Solvers
职业:在线性和混合整数规划求解器中实现鲁棒数值保证的理论和计算进展
  • 批准号:
    2340527
  • 财政年份:
    2024
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Continuing Grant
CAREER: Formal Guarantees for Neurosymbolic Programs via Conformal Prediction
职业:通过保形预测对神经符号程序提供正式保证
  • 批准号:
    2338777
  • 财政年份:
    2024
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Continuing Grant
CAREER: Dual Reinforcement Learning: A Unifying Framework with Guarantees
职业:双重强化学习:有保证的统一框架
  • 批准号:
    2340651
  • 财政年份:
    2024
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Continuing Grant
Collaborative Research: FMitF: Track I: DeepSmith: Scheduling with Quality Guarantees for Efficient DNN Model Execution
合作研究:FMitF:第一轨:DeepSmith:为高效 DNN 模型执行提供质量保证的调度
  • 批准号:
    2349461
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Standard Grant
Theoretical Guarantees of Machine Learning Methods for High Dimensional Partial Differential Equations: Numerical Analysis and Uncertainty Quantification
高维偏微分方程机器学习方法的理论保证:数值分析和不确定性量化
  • 批准号:
    2343135
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Standard Grant
Collaborative Research: Coordinating Offline Resource Allocation Decisions and Real-Time Operational Policies in Online Retail with Performance Guarantees
协作研究:在绩效保证下协调在线零售中的线下资源分配决策和实时运营策略
  • 批准号:
    2226901
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Standard Grant
Collaborative Research: Coordinating Offline Resource Allocation Decisions and Real-Time Operational Policies in Online Retail with Performance Guarantees
协作研究:在绩效保证下协调在线零售中的线下资源分配决策和实时运营策略
  • 批准号:
    2226900
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Standard Grant
CAREER: Towards Tight Guarantees of Markov Chain Sampling Algorithms in High Dimensional Statistical Inference
职业:高维统计推断中马尔可夫链采样算法的严格保证
  • 批准号:
    2237322
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Continuing Grant
Performance Guarantees for Electric Vehicle Fast Charging Station Management
电动汽车快速充电站管理的绩效保证
  • 批准号:
    2312196
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Standard Grant
RI: Medium: Techniques for Massive-Scale Strategic Reasoning: Imperfect-Information Subgame Solving and Offering Guarantees in Simulation-Based Games
RI:中:大规模战略推理技术:不完美信息子博弈解决并在模拟游戏中提供保证
  • 批准号:
    2312342
  • 财政年份:
    2023
  • 资助金额:
    $ 20.5万
  • 项目类别:
    Standard Grant
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了