Vibhhu Sharma

Welcome to my page!

I’m a Master’s student in Machine Learning at Carnegie Mellon University, where I am working with Prof. Bryan Wilder on problems in causal inference and algorithmic decision-making for social impact. Prior to this, I completed my Bachelors in Electrical Engineering from IIT Madras in 2023. My research aims to use machine learning to aid human decision-making in a principled manner that is data-driven, causally motivated, and avoids ethical harm.

I’ve been fortunate enough to complete some very rewarding internships at Instacart, where I worked on improving the coverage of the sequence recommendation model, and Adobe Research, where I developed a model to counterfactually explain the recommendations provided by a multimodal recommender system. More details about that work here!

I worked with Prof. Leqi Liu and Prof. Zachary Lipton at CMU to develop a unified causal framework for auditing recommender systems: viewing recommender system auditing from a causal lens and providing a general recipe for defining auditing metrics. While at IIT Madras, I worked under Prof. Mitesh Khapra and tackled the task of generating diverse natural language counterfactuals for over 11 different Indic Languages. I also interned with Prof. Rohit Babbar at Aalto University and worked on developing deep learning models for Extreme Multilabel Classification.

Apart from academics, I’m a passionate basketball fan and spend a large chunk of my free time watching games. #RipCity

I also write on occasion: you can find some of my recent ramblings here and some of my older work here. I was the Executive Editor at The Fifth Estate, the official campus news body of IIT Madras, where I oversaw a team of 48 people and the daily roll out of content.