Name: Sandeep Madireddy
Pronouns:
Biography:
Sandeep Madireddy is an Assistant Computer Scientist in the Mathematics and Computer Science Division at Argonne National Laboratory. His research interests span the broader areas of theoretical and applied machine learning, probabilistic modeling and high performance computing, with applications across science and engineering. His current research aims at developing foundation models, deep learning algorithms and architectures tailored for scientific machine learning, with a particular focus on improving training efficiency, model robustness, uncertainty quantification and feature representation learning. He has experience applying these approaches to address diverse problems in various domains, ranging from physical sciences (material science, high energy physics, climate science) to computer systems modeling and neuromorphic computing.
Institution/Lab: Argonne National Laboratory
Website: http://www.mcs.anl.gov/~smadireddy/
SRP Collaboration Topic/Title: Probabilistic Machine Learning for Scientific Data
Field or research area: Scientific Machine Learning
Please select all the topical areas that apply to your project:
Machine Learning and AI
Brief Abstract:
ome unique challenges in scientific data that needs to be considered while building data-driven models are: (1) noise and uncertainty, (2) data scarcity, and (3) large feature spaces. Probabilistic models are a natural choice to address many of these challenges and provide a systematic approach to reason about the prediction uncertainty. Historically, the adoption of probabilistic modeling approaches has been limited by the scalability of the inference approaches. With the recent advances in Bayesian deep learning, approximate inference approaches and their information-theoretic connections have enabled inference on large-scale (with millions of parameters) models efficiently with modest computational requirements, obtaining state-of-the-art predictive accuracy. This project would involve building such probabilistic deep learning models for applications such as equilibrium reconstruction of plasma profiles in a magnetically confined fusion tokamak and detecting strong gravitational lenses from astronomical observations from telescopes.
Desired relevant skills, background, or interests:
Python programming, applied mathematics/statistics
Other comments:
Do any special requirements apply? other
Other, specify: None
Keywords:
Machine learning, Scientific Machine Learning, Uncertainty Quantification, Probabilistic,
Lightning Talk Title: Scalable Probabilistic AI for Scientific Machine Learning