01
Scientific Machine Learning
Physics-Informed Neural Networks (PINNs), DeepONets, and neural operators for solving PDEs, inverse problems, and multiphysics systems.
SPARKS Lab
Texas State University · Computer Science
Scientific Prediction through AI Research, Knowledge & Simulation
Developing physics-grounded artificial intelligence and scientific machine learning methods for prediction, discovery, optimization, and engineering design.
About the Lab
AI grounded in scientific principles
The SPARKS Lab (Scientific Prediction through AI Research, Knowledge & Simulation) develops next-generation AI methods for scientific and engineering systems. Our goal is to build machine learning models that do more than fit data: they incorporate physical structure, scale to complex systems, and provide useful representations for scientific discovery and decision-making.
Our research spans physics-informed learning, neural operators, generative AI for science, and hybrid physics–ML modeling, with applications in climate and Earth systems, turbulence, nanoscale heat transport, inverse design, and engineering optimization.
Research
Research areas
We develop computational methods at the intersection of machine learning, applied mathematics, physics, and high-performance scientific computing.
02
Climate & Earth System Modeling
Neural-operator bias correction, nudging strategies for E3SM, and hybrid AI–physics approaches for weather and climate prediction.
03
Turbulence & Fluid Dynamics
Generative and diffusion-based operator models for forecasting, super-resolution, and sparse reconstruction of turbulent flow fields.
04
Nanoscale Heat Conduction
Learning-based methods for ultrashort-pulsed laser heating, two-temperature models, and thermal transport in multilayer thin-film systems.
05
Neural Operators & Spectral Learning
High-frequency representation, spectral-bias mitigation, multi-fidelity learning, and operator learning for complex physical systems.
06
Engineering & Inverse Design
Physics-informed optimization for thermal systems, risk-aware routing, mechanical metamaterials, and inverse design using neural operators.
People
SPARKS Lab team
Principal Investigator
Dr. Aniruddha Bora
Assistant Professor of Computer Science
Texas State University
Ph.D., Louisiana Tech University
Former Postdoctoral Research Associate, Brown University
Ph.D., Louisiana Tech University
Former Postdoctoral Research Associate, Brown University
Graduate Students
Christopher M. Coovrey
Ph.D. Student
Department of Computer Science
Texas State University
Texas State University
Collin Reisman
Ph.D. Student
Department of Computer Science
Texas State University
Texas State University
Keerthana Sunil
Ph.D. Student
Department of Computer Science
Texas State University
Texas State University
Undergraduate Researchers
Pawan Pradhan
Undergraduate Researcher
Mechanical Engineering
Texas State University
Texas State University
Arjun Gyawali
Undergraduate Researcher
Computer Science
Texas State University
Texas State University
Prakriti Gautam
Undergraduate Researcher
Computer Science
Texas State University
Texas State University
Alumni & Past Mentees
Sotos Lois — Imperial College London, 2022–2023 · Mathematical finance using PINNs and operator learning
Scholarship
Selected publications
2025
Integrating Neural Operators with Diffusion Models Improves Spectral Representation in Turbulence Modeling
Proceedings of the Royal Society A
2025
Characterization and Inverse Design of Stochastic Mechanical Metamaterials Using Neural Operators
Advanced Materials
2025
XAI4Extremes: An interpretable ML framework for understanding extreme-weather precursors
ICLR 2025 Workshop
2023
Learning bias corrections for climate models using deep neural operators
AAAI 2023
2022
Neural network method for solving nonlocal two-temperature nanoscale heat conduction in gold films
International Journal of Heat and Mass Transfer
Support & Infrastructure
Funding and computing resources
PIER: Physics-Informed, Energy-efficient, Risk-aware Routing
Texas State University research support for physics-informed and data-driven methods for intelligent routing and decision-making.
$12,000 · 2026–Present
ALCF Director's Discretionary Allocation
High-performance computing allocations supporting physics-informed generative AI and extreme-weather modeling with neural operator methods.
Argonne Leadership Computing Facility
MURI Program (ONR)
Machine-learning methods for phase-change heat-transfer modeling and design, with contributions developed during work at Brown University.
Research Contributor
Leadership-Class Computing
SPARKS research uses ALCF Polaris and Aurora, along with Brown University's OSCAR cluster, for large-scale scientific machine learning experiments.
Polaris · Aurora · OSCAR
Opportunities
Join the SPARKS Lab
We welcome motivated graduate and undergraduate researchers interested in developing rigorous machine-learning methods for scientific and engineering systems. Students with backgrounds in computer science, applied mathematics, physics, and engineering are encouraged to get in touch.
Scientific Machine Learning Physics-Informed AI Neural Operators Generative AI for Science Computational Modeling
Contact Dr. Bora Contact
Get in touch
Dr. Aniruddha Bora
Department of Computer Science
310D COMAL, Texas State University
San Marcos, TX 78666
Department of Computer Science
310D COMAL, Texas State University
San Marcos, TX 78666