About
I am a Senior Data Scientist in Machine Learning at Walmart Global Tech and a computer science Ph.D. graduate from Virginia Tech's Algorithms Lab, where I worked with Sharath Raghvendra. My research interests center on designing scalable algorithms for optimal transport, semantic retrieval, and generative recommendation systems. Beyond my doctoral work, I earned an M.S. in Mining and Mineral Engineering at Virginia Tech advised by Bahareh Nojabaei, with a focus on numerical simulation of flow transportation in complex multiphase reservoir systems.
My work has been published in leading machine learning venues, including ICML, NeurIPS, and ICLR, and spans topics ranging from partial transport metrics to large-scale retrieval and post-training pipelines for generative models.
Selected Projects
I worked in the Algorithms Lab at Virginia Tech on theory-inspired machine learning. Recent projects include:
- Designed robust partial Wasserstein metrics that enable prior-free positive-unlabeled learning and distribution comparison under shift.
- A New Robust Partial p-Wasserstein-Based Metric for Comparing Distributions (ICML 2024). paper · code (GitHub)
- Computing all Optimal Partial Transports (ICLR 2023). paper · code (GitHub)
- Built graph-based parallel algorithms that approximate Wasserstein distance beyond Sinkhorn acceleration on GPUs and large clusters.
- A Combinatorial Algorithm for Approximating Optimal Transport in Parallel and MPC (NeurIPS 2023). paper · code (GitHub)
- Invented automatic delivery route assignment technology during my work at Walmart Global Tech.
- Methods and apparatus for automatic route assignment (U.S. Patent). patent
Academic Service
- Reviewer for NeurIPS, ICML, ICLR, and IJCAI.
Teaching
- CS4104 Data and Algorithm Analysis (Virginia Tech)
- CS3214 Computer Systems (Virginia Tech)
- CS5526 Data Analytics II (Virginia Tech)
Professional Experience
For detailed industry experience, please visit my LinkedIn profile.