NSF HDR DIRSE-IL: I-AIM: Interpretable
augmented intelligence for multiscale material discovery
In this project, we will combine geometric / topological methods with machine learning approaches to help both provide understanding of the principles behind initial structure configurations to final material properties, and build better material (with a focus on composite polymer-carbon nanotubes). Other than me, the team consists of a faculty from Statistics (Johns Hopkins), material science (U. Colorado, Boulder and Illinois Institute of Technology, Chicago), and from Computational mechanics/applied math (Columbia Univ). This is Phase I, and we are hoping to build a foundation so as to support a Phase II institute in the future.
I am hoping to find someone who is interested in applying various geometric / topological ideas to develop new machine learning and data analysis algorithms to help tackle these problems for material science.
If you have anyone that could be good for this (and interested), please let me know, or please ask her/him/them to contact me (email@example.com).