News & Milestones

Our preprint TokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems is now available on arXiv.

My joint work with Mohammed Adnan, Rohan Jain, Tom Jacobs, Rahul G. Krishnan, Rebekka Burkholz, and Yani Ioannou on SparseOpt: Addressing Normalization-induced Gradient Skew in Sparse Training is accepted to ICML 2026.

Ph.D. in Computer Science conferred at the University of Toronto. Thesis: Exploiting linear connectivity modulo permutations in neural networks for parsimonious machine learning.

Started at Google as a Machine Learning Engineer working on large-scale recommender systems for YouTube Shorts.

My joint work with Mohammed Adnan, Rohan Jain, Rahul G. Krishnan, and Yani Ioannou on Sparse Training from Random Initialization: Aligning Lottery Ticket Masks using Weight Symmetry is accepted to ICML 2025.

Entered the job market following my Ph.D. research (archived milestone — see my updated CV).

My joint work with Mohammed Adnan, Rohan Jain, and Yani Ioannou on Winning Tickets from Random Initialization: Aligning Masks for Sparse Training is accepted to the UniREPS workshop at NeurIPS 2024.

New preprint on The Non-Local Model Merging Problem: Permutation Symmetries and Variance Collapse is out on arXiv, with code open-sourced in tact-merge.

My joint work with Devin Kwok, Tom Denton, Daniel M. Roy, David Rolnick, and Gintare K. Dziugaite on Simultaneous Linear Connectivity of Neural Networks Modulo Permutation is accepted to ECML PKDD 2024.

Our work on Simultaneous Linear Connectivity of Neural Networks Modulo Permutation was accepted for a non-archival spotlight at CPAL 2024.

Transitioned my primary research focus from probabilistic programming to deep learning optimization and loss-landscape geometry. Thank you, Dan, for your support!

Moved to Edinburgh for a few months to intern as an Applied Scientist on the Sponsored Display team at Amazon.com under Ben Allison.

My joint work with Zacharie Naulet, Daniel M. Roy, and Victor Veitch on Bootstrap estimators for the tail-index and for the count statistics of graphex processes is accepted to the Electronic Journal of Statistics.

My joint work with Daniel M. Roy on Approximations in Probabilistic Programs: A Compositional Nonasymptotic Analysis of Nested MCMC is accepted to ProbProg 2020.

Presented a poster at the Program Transformations for ML workshop at NeurIPS 2019.

Transitioned from the M.Sc. program to the Ph.D. program at the University of Toronto.

Gave a talk at the Probabilistic Programming Languages, Semantics, and Systems (PPS 2018) workshop at POPL 2018.