In many modern computational and data science applications, novel and efficient numerical techniques are needed to tackle challenges posed by complex physical models and massive datasets. Motivated by the need to address these challenges, this program will focus on three overlapping themes: (i) Randomized Numerical Linear Algebra (RandNLA) algorithms, (ii) Global Sensitivity Analysis, and (iii) Inverse Problems and Uncertainty Quantification, and (iv) Dimensionality Reduction. The goal of this program is to address foundational questions by bringing together researchers in numerical analysis, theoretical computer science, scientific computing, machine learning, statistics, and domain experts. Several working groups will be launched (virtually) in late August 2020, and the aim is to hold two workshops in Spring 2021. Activities in this program will be complemented by the NSF-Funded research training grant titled ‘RTG: Randomized Numerical Analysis’.
- Working Group I: Large-scale Inverse Problems and Uncertainty Quantification (Leaders – Julianne Chung, VT and Arvind K. Saibaba, NCSU)
- Working Group II: Global Sensitivity Analysis (Leaders – Pierre Gremaud, NCSU and Thierry Klein, ENAC & Institut de Mathématiques de Toulouse, University of Toulouse)
- Working Group III: Randomized Algorithms for Matrices and Data (Leader – Arvind K. Saibaba, NCSU)
- Working Group IV: Computational Algorithms for Reinforcement Learning (Leader – Rui Song, NCSU)
- Working Group V: Dimension Reduction in Time Series (Leader – Yaser Samadi, Southern Illinois University)
Questions: email email@example.com