Sourav Dutta

prof_pic.jpg

I am currently a Research Associate in the Computational Hydraulics Group at the Oden Institute for Computational Engineering & Sciences of the University of Texas at Austin, working with Prof. Clint Dawson. Previously, I was an ORISE postdoctoral fellow at the Coastal & Hydraulics Laboratory of the U.S. Army Engineer Research and Development Center, where I worked with Dr. Matthew Farthing. I received my Ph.D. in Mathematics from Texas A&M University in 2017, supervised by Prof. Prabir Daripa.

My research interests lie at the intersection of classical, physics-based computational methods and modern data-driven, machine learning-based techniques with applications to computational science and engineering. I am particularly interested in exploring ways to develop efficient and robust numerical approximations of real-world, large scale environmental flow problems by combining physical principles with modern machine learning algorithms, either by infusing physics-based regularization in the learning trajectory or by modeling the underlying differential operator.

Interests
  • Scientific Machine Learning
  • Model Order Reduction
  • Computational Hydrology
  • Uncertainty Quantification
Education
  • Ph.D. in Mathematics, 2017

    Texas A&M University

  • B.Sc. & M.Sc. in Mathematics & Computing, 2010

    Indian Institute of Technology Kharagpur

Work Experience

Oden Institute for Computational Engineering & Sciences, UT Austin

Research Associate

Mar 2025 - Present · Mississippi (Remote)

Research Fellow

Oct 2022 - Feb 2025 · Mississippi (Remote)

U.S. Army Engineer Research & Development Center (ERDC)

ORISE Postdoctoral Fellow

Sep 2017 - Aug 2022 · Mississippi

News

Aug 01, 2026 Delighted to announce a new 3-year award from the National Science Foundation’s AI and Geosciences program, for the project “Collaborative Research: CAIG: Improving the understanding of coastal groundwater and ocean exchange using machine learning.” The project is led by Dr. Jonghyun Harry Lee (University of Hawai’i at Mānoa) as PI, and I will serve as co-PI leading the UT Austin team on developing neural-operator-based surrogate models for coastal groundwater–ocean exchange, including seawater intrusion and submarine groundwater discharge. Award runs September 2026 through August 2029.
May 01, 2026 Attending the HydroML 2026 Symposium during May 19-21, 2026, at the Oden Institute, and presenting our work on developing MITONet, a neural operator emulator for shallow water flows as well as related DeepONet-based surrogate models for spectral wave models. Also, attending USNC-TAM26, the 20th U.S. National Congress on Theoretical and Applied Mechanics, in Pasadena California during June 21-25, 2026, and presenting some preliminary results of an ongoing study on developing neural operator emulators for salt water intrusion in coastal aquifers.
Oct 01, 2022 I have started my new position as Research Fellow at the Oden Institute for Computational Engineering & Sciences of The University of Texas at Austin. I will be working with Prof. Clint Dawson and other esteemed members of the Computational Hydraulics Group. Looking forward to an exciting and productive time ahead.
Sep 01, 2022 Co-organizing a minisymposium on Machine Learning and Data-Driven Methods for Forward and Inverse Problems along with Matthew Farthing and Dhruv Patel at the SIAM Mathematics of Data Science 2022 meeting in San Diego, CA. Session 1 Session 2
Jul 01, 2022 Giving a talk on Physics-Aware Machine Learning Model for Predicting Coastal Hydrodynamics at the SIAM Annual Meeting 2022 in Pittsburgh, PA.

Selected Publications

  1. JGRML
    Preview of A Neural Operator Emulator for Coastal and Riverine Shallow Water Dynamics
    A Neural Operator Emulator for Coastal and Riverine Shallow Water Dynamics
    Peter Rivera-Casillas, Sourav Dutta, Shukai Cai, and 7 more authors
    Journal of Geophysical Research: Machine Learning and Computation, May 2026
  2. JCP
    Preview of A greedy non-intrusive reduced order model for shallow water equations
    A greedy non-intrusive reduced order model for shallow water equations
    Sourav Dutta, Matthew W. Farthing, Emma Perracchione, and 2 more authors
    Journal of Computational Physics, Aug 2021
  3. MCA
    Preview of Reduced Order Modeling Using Advection-Aware Autoencoders
    Reduced Order Modeling Using Advection-Aware Autoencoders
    Sourav Dutta, Peter Rivera-Casillas, Brent Styles, and 1 more author
    Mathematical and Computational Applications, Apr 2022
  4. SI
    Preview of pyNIROM—A suite of python modules for non-intrusive reduced order modeling of time-dependent problems
    pyNIROM—A suite of python modules for non-intrusive reduced order modeling of time-dependent problems
    Sourav Dutta, Peter Rivera-Casillas, Orie M. Cecil, and 1 more author
    Software Impacts, Aug 2021