Bio

I am a Master's student in Computer Science at Columbia University. I received my Bachelor's in Mathematics and Computer Science from the University of California, Los Angeles. My research interests lie in methods for generative modeling, building faster, more controllable, and more reliable generative models, and generative models for science, leveraging generative models to dramatically accelerate the pace of scientific progress.

I'm grateful to have been advised during my undergrad by Prof. Kim-Lien Nguyen on machine learning for MR physics at the Cardiovascular Imaging Research Lab at the David Geffen School of Medicine, where I worked on k-space (the Fourier, or spectral, domain) reconstruction across two directions: physics-informed reconstruction and 5D MRI reconstruction. I was also fortunate to work with Prof. Yuchen Cui on zero shot imitation learning at the Robotic Intelligence Lab. My research was funded by the James T. Henry and Meredith Henry UCLA Undergraduate Research Scholarship. I have previously worked at AWS Athena, Amazon Leo, Leidos AI/ML Research Accelerator, Scale AI, and Q.ai (Acquired by Apple).

Publications

KARP: K-space Accelerated Reconstruction with Physics Priors
Thomas Coudert*, Amit Rand*, Zhengyang Ming, Xinran Gao, Dan Ruan, Kim-Lien Nguyen
2026 IEEE Transactions on Biomedical Engineering. Under review.
MCCVDM architecture diagram
Complex-valued diffusion in k-space for accelerated MRI reconstruction
Amit Rand, Thomas Coudert, Zhengyang Ming, Xinran Gao, Dan Ruan, Kim-Lien Nguyen
Preprint
5D free-running CMR reconstruction
Diffusion-based k-space inpainting for improved 5D free-running CMR reconstruction
Thomas Coudert, Amit Rand, Xinran Gao, Zhengyang Ming, J. Paul Finn, Dan Ruan, Kim-Lien Nguyen
2026 ISMRM & ISMRT Annual Meeting & Exhibition. Cape Town, South Africa
MXA architecture diagram
Beyond Conventional Transformers: A Medical X-ray Attention Block for Improved Multi-Label Diagnosis
Amit Rand, Hadi Ibrahim
2025 NeurIPS, Imageomics: Discovering Biological Knowledge from Images Using AI. San Diego, California, USA