Research

As explained on my philosophy page, I care about utilizing interdisciplinary tools that I know (statistics, optimization, and analytics) to create impact. I work on both methodology and applications to achieve this goal across my core focus areas:

Optimal and Scalable Machine Learning

Machine learning is already very powerful, but there are still so many tasks outside of its reach. How do we develop machine learning and optimization algorithms for difficult, large-scale tasks? This stream of work aims to push the boundaries on algorithm tractability while providing performance guarantees.

Published in Journal of Machine Learning Research, INFORMS Journal of Computing, and more.

Safe Evaluation of Machine Learning

How can we make sure a machine learning algorithm is doing what we want it to do, at the performance of what we expect? We develop new methodologies for rigorous and safe evaluation of machine learning under both randomized and observational settings with minimal assumptions.

Published in Journal of the American Statistical Association, Manufacturing & Service Operations Management, and more.

Applications in Healthcare

The best test of any method is practice. I apply my methodology in hospitals, pharmaceutical companies, and organizations worldwide to create safe personalized treatments for pediatric patients, optimized clinical trials for life-saving vaccines, mitigation plans for governmental restrictions, and more.

Published in Operations Research, Proceedings of the National Academy of Sciences and more.

Other
Applications

Beyond my core focus in healthcare-related areas, I also have a wider interest in other important applications of machine learning and optimization, with a particular interest in supply chain.

Published in International Conference on Learning Representations and more.