Somjit Roy

At Oppenheimer's House

I am currently a doctoral student in Statistics at Texas A&M University, where I am advised by Dr. Bani K. Mallick (TAMU Statistics) and co-advised by Dr. Debdeep Pati (UW-Madison Statistics).

My research focuses on:

Developing statistical modeling frameworks that unify data-driven learning and scientific reasoning to solve complex, real-world problems in materials discovery, geosciences, physics, and computational genomics.
Integrating Scientific machine learning and Bayesian modeling & computation through scientifically-guided probabilistic inference, physics-informed modeling, variational inference, tree-based models, and Bayesian optimization.

For more insights into my work, visit the Research page.

Oct 28, 2025	Talk on Bayesian Symbolic Regression for Structural Learning of Scientific Expressions at STAT CAFÉ , TAMU Statistics, College Station, TX.
Sep 27, 2025	Invited talk on SPINWAVE at the minisymposium on Computational Methods for Multiscale and Multiphysics Problems at the 8th Annual SIAM Texas–Louisiana Workshop 2025 , Austin, TX.
Sep 24, 2025	Preprint on Bayesian SciML for Structural Learning of Scientific Expressions, submitted.
Sep 24, 2025	First official GitHub release of HierBOSSS on .
Jul 27, 2025	First official GitHub release of TAVIE on .
May 19, 2025	Graduate Intern in the Earth and Environmental Sciences (EES) division at Los Alamos National Laboratory during the summer of 2025.
Apr 07, 2025	Preprint on tangent-transform variational inference for strongly super-Gaussian likelihoods, .
Dec 28, 2024	Talk on TAVIE: Tangent Approximation for Variational Inference in different Exponential Families at IISA 2024.
Nov 19, 2024	Paper on Sparse Almost Perfect Mutually Unbiased Bases (APMUBs) got published in Journal of Statistical Theory and Practice, .