Internship 3
Host name: Bianca Dumitrascu
School and Department: Department of Statistics, Irving Institute for Cancer Dynamics (IICD)
Internship title: Physical AI for Lung Development
Number of interns to be hosted: up to 2
Types of support offered:
- Stipend: up to $14,000 over the duration of the internship
- Access to Columbia University campus services, computational resources, and lab meetings
- Mentorship and supervision by faculty and senior lab members
- Immigration and Visa assistance through Columbia’s International Students and Scholars Office (ISSO)
Internship description
Is it a lung or is it a salivary gland? Many organs look like trees, but is it the same type of tree? Our lab has developed physical AI emulators for studying how cells build spatio-temporal structures. You will be using these systems to simulate trees emerging from PDEs with different boundary conditions. You will also develop methods for PDE constrained graph neural networks that fit real data to learn why our organs look the way they do. You will learn about microscopy data, spatial transcriptomics, hybrid physics+AI systems and nonequilibrium thermodynamics.
Skills required
familiarity with pytorch and Python, interest in developmental biology OR multi agent RL systems