Channel Simulation • Information Theory • Machine Learning

I am a 5th year PhD candidate in the school of Electrical and Computer Engineering at Cornell University, where I am grateful to be advised by Aaron Wagner. Previously, I worked as an analyst in the Risk division of Goldman Sachs and received my undergraduate education at the Indian Institute of Technology at Hyderabad. 

My research is focused on proving theoretical bounds and designing practical algorithms for channel simulation, a generalized version of lossy compression without quantization that is closely connected to relative entropy coding . This has several potential applications in neural network-based data compression, federated learning, and differential privacy. However, this potential cannot be realized yet because of the prohibitive scaling complexity of the known algorithms. My goal is to address this.

I have worked on both the theoretical and practical sides of channel simulation, deriving state-of-the-art bounds on the fundamental limits of point-to-point simulation of i.i.d. channels (recognized with the Jack Keil Wolf Student Paper Award at ISIT 2024 and as a finalist for the award in 2025)  and developing fast and scalable algorithms for simulating a limited class of channels. 

In my free time, I enjoy playing some recreational sports, mainly pool, chess, and table tennis. Feel free to challenge me on lichess if you want to play some blitz!