Research

I work on probabilistic AI and AI for healthcare. Specifically, I am interested in developing methods to reconcile information from multi-source data with different forms of domain knowledge into human-understandable forms. Since this scenario is frequently encountered in knowledge-rich but data-poor domains such as healthcare, I am also interested in adapting these methods for complex clinical problems such as modeling the risk of adverse pregnancy outcomes (e.g., preterm birth and gestational diabetes) and understanding the causes of neurological injury in children on life-support (ECMO).

To this effect, my work has considered clinical domain knowledge about relationships between two or more variables including causal influence, monotonicities, and independence. We have developed methods to combine these diverse forms of domain knowledge with noisy, sparse, and uncertain data to construct explainable and interpretable probabilistic graphical models including Bayesian and credal networks, tractable probabilistic models such as sum-product networks and cutset networks, and neurosymbolic models.

Before joining UTD, I was a Master’s student at Indiana University. I worked with Prof. David Crandall on bayesian uncertainty quantification algorithms for deep neural networks. I completed my M.S. in May 2020.

As an undergraduate student at Vellore Institute of Technology, I worked with Prof. Daphne Lopez on deep models for image caption generation and chatbots. I have interned at Microsoft, Bangalore during my undergrad and Synopsys, Mountain View during my masters.

Publications

Teaching

  • University of Texas, Dallas

    University of Texas, Dallas

    Teaching Assistant

    • CS6375 - Machine Learning (Spring 2021)
    • CS4375 - Introduction to Machine Learning (Fall 2020)
    • CS6301 - Software Defined Networking (Fall 2020)
  • Indiana University, Bloomington

    Indiana University, Bloomington

    Associate Instructor

    • B657 - Computer Vision (Spring 2020)
    • B551 - Elements of AI (Fall 2019)
    • B456 - Image Processing (Spring 2019)
  • IEEE Computer Society, VIT Chapter

    IEEE Computer Society, VIT Chapter

    Workshop Instructor

    • Game Development with HTML5 (Fall 2016)
    • Python Programming (Fall 2016)
    • C++ Programming (Spring 2016)

Work Experience

  • Synopsys, Mountain View

    Synopsys, Mountain View

    R&D Intern

    • Developed a neural machine translation system to generate assertions from specifications.
  • Indiana University, Bloomington

    Indiana University, Bloomington

    Research Assistant

    • Visualized 94,000 geotagged Irish folktales using Bayesian topic models under Prof. Roni Khardon.
  • Microsoft, Bangalore

    Microsoft, Bangalore

    Summer Intern

    • Built a movie recommendation engine with an error-rate of 0.87 on the Netflix dataset.

Projects

Causal Bayesian Network construction for patients on ECMO using Large Language Models

Causal Bayesian Network construction for patients on ECMO using Large Language Models

Pacific Symposium on Biocomputing, 2025

Constructing causal graphs to understand the risk of neurological injury for patients on life support by refining an LLM's output using real clinical data and domain knowledge.

Speech to text

Speech to text

GitHub, 2019

Deep learning based end-to-end speech recognition system

Semantic and Instance Segmentation

Semantic and Instance Segmentation

GitHub, 2019

Deep learning and PGM based image segmentation

Automated Image Captioning System

Automated Image Captioning System

GitHub, 2018

Deep learning based image caption generator to improve accessibility on the web.

VITacademics

VITacademics

GitHub, 2018

A web-server for mobile applications that streamline access to academic information for students at VIT.

Clickbait Detector

Clickbait Detector

GitHub, 2018

Detecting clickbait headlines in the wild with 90% accuracy.

Movie Recommendation Systems

Movie Recommendation Systems

GitHub, 2016

A comparitive study of collaborative filtering algorithms.

graVITas, 2016

graVITas, 2016

GitHub, 2016

Web-portals for VIT University’s annual technical festival.