About me

I’m an AI Research Scientist at Ground Truth Analytics (GTA) working on deep learning based satellite imagery for agricultural applications. I recently graduated from Rice University ECE department. There, I worked in the Machine Learning and Information processing Laboratory (MLI Lab) and was very lucky to be advised by Reinhard Heckel. Before that, I did my undergrad in EE at Sharif University of Technology.

I’m highly passionate about AI for Good. In particular, I enjoy studying the performance and reliability of Deep Learning models in computer vision applications such as image classification and image reconstruction.

News

• [Aug 2023] Research: New paper “IR-FRestormer: Iterative Refinement with Fourier-Based Restormer for Accelerated MRI Reconstruction” accepted to WACV 2024.
• [May 2023] Work: I joined GTA to work with a brilliant group to revolutionize agriculture via satellite imagery. We work on a variety of data-driven approaches to enhance global food security.
• [May 2023] Research: I successfully defended my Ph.D. thesis!
• [May 2022] Research: New paper “Test-time training can close the natural distribution shift performance gap in deep learning based compressed sensing” accepted to ICML 2022 [paper, Code].
• [May 2022] Research: I’ll join Nvidia Research as an Applied Research Scientist Intern for Summer 2022; very excited about it!
• [Jun 2021] Research: New paper “Accelerated MRI with un-trained neural networks” accepted to IEEE Transactions on Computational Imaging [paper, code].
• [Jun 2021] Invited talk: Guest speaker at UCSF Journal Club for robustness of deep learning based compressed sensing methods.
• [May 2021] Research: Our interactive demo for accelerated MRI reconstruction using un-trained networks won the 2nd place in the MR-Pub Competition.
• [May 2021] In the media: Our work on the robustness of deep learning based image reconstruction methods appeared on Stanford news.
• [May 2021] Research: New paper “Measuring robustness in deep learning based compressive sensing” accepted to ICML 2021 as a long talk (top 3% papers) [paper, code].
• [Feb 2021] Research: New abstract “Can un-trained networks compete with trained ones for accelerated MRI?” accepted to ISMRM 2021 as an oral presentation at the “Machine Learning for Image Reconstruction” session [paper, code].
• [Oct 2020] Research: New paper “Low cost DNA data storage using photolithographic synthesis and advanced information reconstruction and error correction” accepted to Nature Communications [paper, code].