TRIPODS+X:RES: Collaborative Research: Creating Inference from Machine Learned and Science Based Generative Models

TRIPODS X:RES：协作研究：从机器学习和基于科学的生成模型中创建推理

• 批准号: 1839217
• 负责人: Uros Seljak
• 金额: $ 39.94万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1839217&HistoricalAwards=false
• 关键词: TRIPODS+X; RES; Collaborative; Research; Creating

In many scientific disciplines computational simulations are used to enhance our understanding of physical processes of complex systems. In all such simulations simplifications are required to make the problem tractable, limiting the scope of the questions that can be addressed. Generally, computational simulations of large systems with many interacting components, based on the governing physics, requires complex and time consuming computations. This project will apply deep learning neural networks (NN) with geometric transformations based on the physics of the system to accurately approximate traditional physics-based computational simulations in a highly efficient manner. The increased efficiency imparted by the NN model will facilitate the asking of scientific questions which are currently computationally intractable. While the proposed work will focus on using this method to discover new strain-induced polar phases and phase competition, and to understand the large-scale structure in the universe, the concepts developed in this work can be applied to computational simulations in other scientific disciplines.The proposed work will focus on the development of foundational data science methods and the application of these methods to augment computationally-expensive science-based generative models in a way that is principled and efficient, thereby enabling improved data-driven scientific inference. The work will place specific emphasis on the design of neural network models, which through physically-significant domain architectures can approximate N-body and highly-correlated phenomena with minimal loss of information. The work will develop tools to guide the discovery and experimental synthesis of new strain-induced polar phases and phase competition, which exhibit enhanced electromechanical responses; and it will expand our simulation capabilities and understanding of the large-scale structure in the universe. Ultimately, this work will provide both domain specific advances, as well as a framework for other domain areas to augment computationally intensive, highly-correlated, N-body problems with data-driven models, which respect the physics of the problem and lead to increased computational efficiency.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.

在许多科学学科中，计算模拟用于增强我们对复杂系统物理过程的理解。在所有此类模拟中，都需要简化问题来使问题可解决，从而限制了可以解决的问题的范围。通常，基于管理物理学的许多相互作用组件的大型系统的计算模拟需要复杂而耗时的计算。该项目将基于系统物理学的几何变换应用深度学习神经网络（NN），以高效的方式准确地近似基于物理的计算模拟。 NN模型赋予的效率提高将有助于提出目前计算上棘手的科学问题。虽然拟议的工作将集中于使用这种方法来发现新的应变引起的极性阶段和阶段竞争，并了解宇宙中的大规模结构，但这项工作中开发的概念可以应用于其他科学学科中的计算模拟。拟议的工作将集中于基础数据科学及其在这些方法上的应用，以增强科学效果，并在其上及其实现的方法，以增强其效果，以增强其效果，以增强科学效果，以实现促进效果，并具有促进性的生产力，以实现型模型，并具有良好的生产力，并构成了一种良好的生产力，并构成了一种良好的生产力，并将其用于开发。数据驱动的科学推断。这项工作将特别强调神经网络模型的设计，通过物理上重要的域结构可以近似N体和高度相关的现象，而信息的损失最小。这项工作将开发工具，以指导新应变诱导的极性相和相位竞争的发现和实验综合，这些相位表现出增强的机电反应；它将扩大我们的模拟能力和对宇宙大规模结构的理解。最终，这项工作将提供特定领域的进展，也将为其他领域的领域提供一个框架，以增强计算密集型，高度相关的，具有数据驱动模型的N体问题，这些模型尊重问题的物理和计算效率的提高。该奖项反映了NSF的法定任务，并通过评估范围来反映出支持者的知识群体，并通过评估范围进行了评估和宽广的知识。

项目成果

Translation and rotation equivariant normalizing flow (TRENF) for optimal cosmological analysis

用于最佳宇宙学分析的平移和旋转等变归一化流 (TRENF)

• DOI: 10.1093/mnras/stac2010
• 发表时间: 2022
• 期刊: Monthly Notices of the Royal Astronomical Society
• 影响因子: 4.8
• 作者: Dai, Biwei;Seljak, Uroš
• 通讯作者: Seljak, Uroš

FlowPM: Distributed TensorFlow implementation of the FastPM cosmological N-body solver

• DOI: 10.1016/j.ascom.2021.100505
• 发表时间: 2020-10
• 期刊: Astron. Comput.
• 作者: C. Modi;F. Lanusse;U. Seljak

Kepler Data Analysis: Non-Gaussian Noise and Fourier Gaussian Process Analysis of Stellar Variability

• DOI: 10.3847/1538-3881/ab8460
• 发表时间: 2020-05-01
• 期刊: ASTRONOMICAL JOURNAL
• 影响因子: 5.3
• 作者: Robnik, Jakob;Seljak, Uros
• 通讯作者: Seljak, Uros

Marginal unbiased score expansion and application to CMB lensing

边际无偏分数扩展及其在 CMB 透镜中的应用

• DOI: 10.1103/physrevd.105.103531
• 发表时间: 2022
• 期刊: Physical Review D
• 影响因子: 5
• 作者: Millea, Marius;Seljak, Uroš
• 通讯作者: Seljak, Uroš

High mass and halo resolution from fast low resolution simulations

通过快速低分辨率模拟获得高质量和光晕分辨率

• DOI: 10.1088/1475-7516/2020/04/002
• 发表时间: 2020
• 期刊: Journal of Cosmology and Astroparticle Physics
• 影响因子: 6.4
• 作者: Dai, Biwei;Feng, Yu;Seljak, Uroš;Singh, Sukhdeep
• 通讯作者: Singh, Sukhdeep