Home
I am currently working with the Search Sourcing team in the Sponsored Products org at Amazon. My primary research interests lie at the intersection of optimization and machine learning. I am currently working on providing a practical robust optimization algorithm that guarantees better generalization than stochastic gradient descent (SGD). I have also collaborated on wide variety of problems in deep learning, non-convex optimization, robust optimization, graphical models and semi-supervised learning.
Previously, I received a PhD in Electrical and Computer Engineering department at The University of Texas at Austin, under the guidance of Prof. Sujay Sanghavi as part of the Wireless Networking and Communications Group (WNCG). I received my B.Tech. and M.Tech. (Dual Degree with a specialization in Communications and Signal Processing) from the Electrical Engineering Department at the Indian Institute of Technology, Bombay in 2014.
Latest Updates
(February 22, 2021) Robust Estimation of Tree Structured Markov Random Fields is up on ArXiv
(December 1, 2020) On Generalization of Adaptive Methods for Over-parameterized Linear Regression is up on ArXiv
(June 13, 2020) 'Robust Estimation of Tree Structured Ising Models' is up on ArXiv
(May 31, 2020) 'Negative Sampling with Semi-supervised Learning' accepted to ICML 2020
(April 9, 2020) Blog on Negative Sampling with Semi-supervised Learning: Link to the blog post
(March 10, 2020) Github Code for MKL-SGD paper: Link to the github repository
(January 7, 2020) 'Choosing the Samples with the Lowest Loss Makes SGD Robust' accepted to AISTATS 2020