Name: Emil Constantinescu
Pronouns:
Biography:
I graduated from Virginia Tech in 2008 and I joined Argonne right after that. I am currently a Computational Mathematician at Argonne. I work on numerical PDEs, uncertainty quantification, and machine learning. I work on weather and climate applications, power grid problems, and nuclear physics.
Institution/Lab: Argonne National Laboratory
Website: https://web.cels.anl.gov/~emconsta
SRP Collaboration Topic/Title: Physics-based machine learning and uncertainty quantification
Field or research area: ML, PDEs, numerical simulation, uncertainty quantification
Please select all the topical areas that apply to your project:
Machine Learning and AI
Brief Abstract:
This project aims to advance computational science by integrating numerical methods for solving partial differential equations (PDEs) with machine learning and statistics for uncertainty quantification. Traditional simulation models have limitations in capturing complex behaviors inherent in systems like weather forecasting, climate modeling, power grid management, and nuclear physics. By incorporating machine learning algorithms into the numerical solving of PDEs, we aspire to build more accurate and efficient simulation models. Additionally, we will employ statistical methods for uncertainty quantification to provide more reliable predictions and to understand the range of possible outcomes. The ultimate goal is to develop a unified computational framework that is versatile enough for applications in various disciplines, from predicting severe weather events to optimizing power grid operations and advancing our understanding of nuclear physics phenomena. We want to explore new approaches to modeling complex systems, offering more accurate and reliable predictions for scientific inquiry and practical applications.
Desired relevant skills, background, or interests:
Machine Learning, Python, Numerical simulation of PDEs, uncertainty quantification
Other comments:
Do any special requirements apply? In-Person Only
Other, specify:
Keywords:
Machine learning, numerical analysis, numerical PDEs, physics-informed ML
Lightning Talk Title: Physics-based machine learning and uncertainty quantification