HDR TRIPODS: Illinois Institute for Data Science and Dynamical Systems (iDS2)

HDR TRIPODS：伊利诺伊州数据科学与动力系统研究所 (iDS2)

• 批准号: 1934986
• 负责人: Maxim Raginsky
• 金额: $ 150万
• 依托单位: University of Illinois at Urbana-Champaign
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1934986&HistoricalAwards=false
• 关键词: HDR; TRIPODS; Illinois; Institute; Data

This project establishes the Illinois Institute for Data Science and Dynamical Systems (iDS2) at the University of Illinois at Urbana Champaign (UIUC) that combines research in data science with dynamical systems. iDS2 is composed of researchers in electrical engineering, computer science, statistics and mathematics. Data science already exerts considerable influence over daily life. As the age of autonomous vehicles, smart cities, personalized medicine, and artificial intelligence continues to emerge, both data-driven algorithms and dynamical feedback loops will permeate every aspect of human existence. It is becoming increasingly clear that the study of dynamical systems is at the heart of fundamental challenges arising in data science. Indeed, system dynamics occur in every facet of data science, from data generation, modeling, and inference, all the way to learning and automation. Since the study of dynamical systems cuts across the disciplines of engineering, mathematics, statistics, and computer science, a synthesis of data science and dynamical systems will be a powerful catalyst for further advances in all of these fields.The aim of this project is to initiate this synthesis through iDS2, which will leverage the multidisciplinary expertise and strengths at UIUC in these two fields. The overall vision of iDS2 is to create and foster a transdisciplinary program that breaks the boundaries between the fields of data science and dynamical systems by developing new foundational principles for both fields and by broadening their applicability. The Phase I activities of the institute will focus on four interrelated research themes: (1) data modeling and dynamical systems; (2) sampling, inference, and dynamical systems; (3) algorithm design and dynamical systems; (4) decision-making and dynamical systems. The institute will foster synergies with existing research and education entities at UIUC in order to catalyze and scale up interactions between the TRIPODS communities of electrical engineering, mathematics, statistics, and theoretical computer science, both at UIUC and in the Midwest.This project is part of the National Science Foundation's Harnessing the Data Revolution (HDR) Big Idea activity.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.

该项目在伊利诺伊大学厄巴纳尚潘分校（UIUC）建立了伊利诺伊数据科学和动态系统研究所（iDS 2），将数据科学与动态系统的研究相结合。iDS 2由电气工程、计算机科学、统计学和数学领域的研究人员组成。 数据科学已经对日常生活产生了相当大的影响。随着自动驾驶汽车、智能城市、个性化医疗和人工智能时代的不断涌现，数据驱动算法和动态反馈回路将渗透到人类生存的各个方面。越来越清楚的是，动力系统的研究是数据科学中出现的基本挑战的核心。事实上，系统动力学出现在数据科学的各个方面，从数据生成、建模和推理，一直到学习和自动化。由于动力系统的研究跨越了工程、数学、统计学和计算机科学等学科，数据科学和动力系统的综合将成为所有这些领域进一步发展的强大催化剂。该项目的目的是通过iDS 2启动这种综合，这将利用UIUC在这两个领域的多学科专业知识和优势。iDS 2的总体愿景是创建和促进一个跨学科的计划，通过为这两个领域开发新的基础原则并扩大其适用性，打破数据科学和动力系统领域之间的界限。该研究所的第一阶段活动将集中在四个相互关联的研究主题：（1）数据建模和动态系统;（2）采样，推理和动态系统;（3）算法设计和动态系统;（4）决策和动态系统。该研究所将促进与UIUC现有研究和教育实体的协同作用，以促进和扩大电气工程，数学，统计和理论计算机科学的TRIPODS社区之间的互动，该项目是国家科学基金会利用数据革命（HDR）的一部分。大创意活动。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

ZenoPS: A Distributed Learning System Integrating Communication Efficiency and Security

ZenoPS：集通信效率与安全于一体的分布式学习系统

• DOI: 10.3390/a15070233
• 发表时间: 2022
• 期刊: Algorithms
• 影响因子: 2.3
• 作者: Xie, Cong;Koyejo, Oluwasanmi;Gupta, Indranil
• 通讯作者: Gupta, Indranil

Learning While Scheduling in Multi-Server Systems With Unknown Statistics: MaxWeight with Discounted UCB

• 发表时间: 2022-09
• 作者: Zixi Yang;R. Srikant;Lei Ying

Sample Complexity and Overparameterization Bounds for Temporal-Difference Learning With Neural Network Approximation

• DOI: 10.1109/tac.2023.3234234
• 发表时间: 2021-03
• 期刊: IEEE Transactions on Automatic Control
• 影响因子: 6.8
• 作者: Semih Cayci;Siddhartha Satpathi;Niao He;F. I. R. Srikant

Collaborative Multi-Agent Heterogeneous Multi-Armed Bandits

协作多智能体异构多臂强盗

• 发表时间: 2023
• 期刊: ICML
• 作者: Chawla, R.;Vial, D.;Shakkottai, S.;Srikant, R.
• 通讯作者: Srikant, R.

A Nonconvex Framework for Structured Dynamic Covariance Recovery

• 发表时间: 2020-11
• 期刊: J. Mach. Learn. Res.
• 作者: Katherine Tsai;M. Kolar;Oluwasanmi Koyejo