I am a PhD student at the Vector Institute and the University of Calgary, supervised by Dr. Yani Ioannou and Dr. Rahul Krishnan. My research goal is to make large foundational models, such as LLMs, smaller and efficient. Currently, my work focuses on developing efficient training methods for Sparse Neural Networks. I am also interested in understanding the training dynamics of (Sparse) Neural Networks, weight/permutation symmetries of Deep Neural Networks, and their effect on optimization and developing a better understanding of the loss landscape. Previously, I worked on the Neural Tangent Kernel to better understand Shortcut Learning in Neural Networks. My research is supported by NSERC, Borealis AI, Digital Research Alliance of Canada and Killiam Fellowship. I received my Bachelor’s Degree in Electronics & Electrical Engineering from the Indian Institute of Technology Guwahati and a MASc degree from the University of Waterloo. Feel free to reach out to me if you would like to discuss research or collaborate.

Academic Services:

• Reviewer: ICLR 2021-25, UAI 2022-24, NeurIPS 2022-25, CVPR 2023-25, ICCV 2023-24.
• Co-host of Self-Supervised Learning reading group at Vector Institute. Feel free to reach out if you’re interested in joining the reading group.
• Student mentorship: Rohan Jain (MASc), Tejas Pote (2023 Summer Intern), Nahush Lele (2023 Summer Intern).

(Research) Updates

• May 2025 - Awarded prestigious Killiam Doctoral Fellowship in recognition of research contribution.
• May 2025 - Our paper on Sparse Training got accepted at ICML 2025. DM me if you have any questions or want to collaborate. See you in Vancouver!
• December 2024 - Our workshop proposal got accepted at ICLR 2025. I will be co-organizing a workshop on Sparsity in LLMs, see you in Singapore!
• October 2024 - Received research funding from Digital Research Alliance of Canada (DRAC) to investigate effect of LLM compression on model bias.
• September 2024 - Among 10 students to be awarded Borealis AI Global Fellowship.
• April 2024 - Awarded prestigious NSERC Doctoral Fellowship from the Government of Canada.
• January 2023 - Started PhD.
• September 2022 - Started Research Internship at Borealis AI.
• January 2022 - Paper accepted to Nature Scientific Reports.
• December 2022 - Paper accepted at the Trustworthy AI for Healthcare Workshop - AAAI 2022.
• November 2021 - Paper accepted at the NeurIPS workshop on ‘Self-Supervised Learning - Theory and Practice’.
• October 2021 - Submitted ongoing work on domain-agnostic SSL to NeurIPS workshop on ‘Self-Supervised Learning - Theory and Practice’.
• September 2021 - Graduated from the University of Waterloo.
• June 2021 - Our paper ‘Pay Attention with Focus: A Novel Learning Scheme for Classification of Whole Slide Images’ got accepted to MICCAI 2021
• July 2020 - Our paper ‘Learning Permutation Invariant Representations using Memory Networks’ has been accepted for publication at ECCV 2020.
• April 2020 - My recent work, “Representation Learning of Histopathology Images using Graph Neural Networks,” got accepted to CVPR(W) 2020.
• September 2019 - Joined the University of Waterloo for graduate studies.
• April 2019 - Awarded Vector Scholarship in AI by the Vector Institute, Canada.
• December 2018 - I will be joining University of Waterloo for graduate studies in Fall 2019.
• October 2017 - Awarded Shastri Indo Canadian Research Fellowship 2018.