Pratik Rathore

I am a second year PhD student in the Electrical Engineering department at Stanford University, interested in randomized algorithms for optimization, advised by Madeleine Udell.

Before Stanford, I graduated from the University of Maryland with a BS in Electrical Engineering and BS in Mathematics. As an undergraduate, I held internships at STR, where I conducted research on radar image processing, and Lockheed Martin, where I reviewed and tested computational models for satellites. I also conducted research in number theory, for which I received the Dan Shanks Award from the University of Maryland Math Department.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  Github

profile photo

I'm interested in using randomization to improve the scalability of optimization algorithms, especially in applications to machine learning. Recently, I've thinking about using sketching and low-rank approximation to develop new optimization algorithms for machine learning tasks.

SketchySGD: Reliable Stochastic Optimization via Robust Curvature Estimates
Zachary Frangella, Pratik Rathore, Shipu Zhao, Madeleine Udell

We use techniques from randomized numerical linear algebra to develop an improved stochastic quasi-Newton method.

There are no Cube-free Descartes Numbers with Exactly Seven Distinct Prime Factors
Pratik Rathore

We prove new results regarding the prime factorizations of Descartes numbers, a family of odd spoof perfect numbers.

umd_ece Teaching Assistant, ENEE150, Spring 2021

Template borrowed from Jon Barron.