Project Description

Real-world problems in computational science and engineering that lack closed-form solutions often require the use of expensive numerical solvers taxing CPU/GPU resources substantially both in processing and memory. Even minor changes to the problem’s parameters frequently require running the numerical solver again, adding to the computational expense and time. In this project, we aim to innovate on solutions for modeling the system in terms of stochastic partial differential equations as well as solving them for future prediction. 

Start Date

February 2025

Post Doc Qualifications

The postdoc must have experience in scientific machine learning with preferred publications in top-tier machine learning venues. 

Co-Advisors

Vaneet Aggarwal, vaneet@purdue.edu, IE/ECE, https://engineering.purdue.edu/CLANLabs
Aniket Bera, aniketbera@purdue.edu, CS, https://www.cs.purdue.edu/homes/ab/ 

Bibliography

https://arxiv.org/pdf/2303.10528
https://papers.nips.cc/paper_files/paper/2023/file/2a4179ef39846557e99f6bfac580ea2e-Paper-Conference.pdf
https://papers.nips.cc/paper_files/paper/2023/file/57c30b677add9aa78e1745f0643104d0-Paper-Conference.pdf
https://papers.nips.cc/paper_files/paper/2023/file/6a27ee6f66d13557f15f070274c51721-Paper-Conference.pdf
https://papers.nips.cc/paper_files/paper/2023/file/df54302388bbc145aacaa1a54a4a5933-Paper-Conference.pdf