Amit Rand

I am a final-year undergraduate studying Mathematics and Computer Science at the University of California, Los Angeles.

My research interests lie in the realm of mechanistic machine learning. Across my research experiences, I have repeatedly seen that the most robust systems are the ones that respect the structure of the underlying problem. I believe the next leap in ML will come from models that don't just fit data, but internalize domain structure. My goal is to help build systems that embed these priors directly into learning. I am pursuing a PhD in Computer Science to explore this topic in depth.

I am currently a research assistant at the Robotic Intellgience Lab (URIL) at UCLA, where I work on robot learning, specifically physics-based retrieval augmented imitation learning advised by Professor Yuchen Cui. I am also a research assistant at the Cardiovascular Imaging Reserach Lab (CVIRL) at the David Geffen School of Medicine at UCLA, where I work on accelerated 5D MRI reconstruction using generative modeling, advised by Professor Kim-Lien Nguyen.

Previously, I interned at Amazon, where I developed physics-informed models for thermal calibration of satellite antenna panels. I also worked as a research scientist intern with the Graph AI group at the Leidos AI/ML Research Accelerator. I've also interned at Scale AI (GenAI) and worked as a software engineer at two startups.

Download Resume

Select Publications and Software

Measurement-conditioned diffusion models for k-space inpainting and image space reconstruction to improve 5D (cardiac + respitory time) free-running cardiac MRI reconstruction.

A novel attention mechanism designed specifically for medical X-ray images to improve multi-label diagnosis.

A comprehensive medical image analysis platform developed by CVIRL @ UCLA enabling advanced 2D/3D medical image segmentation, centerline analysis, and volumetric reconstruction with an intuitive multi-viewer interface.

Last updated: February 04, 2026
template here