Vignesh Kothapalli

I am a CS masters student at the NYU Courant Institute of Mathematical Sciences, fortunate to be advised by Prof. Joan Bruna and Prof. Jonathan Niles Weed. My research focuses on understanding the dynamics of feature learning in (graph) neural networks using techniques from random matrix theory and spectral graph theory. Previously, I had the opportunity to develop gradient compression techniques for fast pre-training of transformer models as a part of LinkedIn’s Deep Learning Infrastructure org. I obtained my bachelors degree in Electronics and Communication Engineering from IIT Guwahati and worked at IBM as a software developer before joining NYU.