Sharmila Duppala

I am an Applied Scientist on the Core Matching & Optimization team at Uber. I completed my Ph.D. in the Computer Science department at the University of Maryland, College Park, where I was fortunate to be co-advised by professors Aravind Srinivasan and John P. Dickerson. My Ph.D. thesis, titled "Algorithmic Fairness and Online Decision Making in Combinatorial Optimization and Unsupervised Learning," focused on designing fair algorithms with provable guarantees for classical combinatorial and clustering problems. Prior to that, I was a Graduate Research Assistant at the Computer Science department at Stony Brook University, where I was very lucky to be advised by Prof. Rezaul Chowdhury. My Master's thesis focused on the reusable resource-allocation problem.

I am broadly interested in algorithmic fairness for combinatorial problems, particularly in the design and analysis of fair algorithms for a wide variety of classical problems like graph matching, set covering, set packing, and clustering. My past research has focused on incorporating probabilistic generalizations of group and individual fairness into combinatorial problems and designing algorithms with provable guarantees.

My ongoing work extends this line of research to combinatorial optimization and machine learning under uncertainty, where the information available about the problem instance — group memberships, item valuations, or user preferences — is stochastic, incomplete, or unreliable. This includes designing matching and clustering algorithms that remain robust when membership or preference data is uncertain, and studying market mechanisms like barter exchange under asymmetric or privately held valuations.

Sharmila Duppala
sduppala AT umd DOT edu

News

Community Service

2018–Present Reviewer
AAAI (2026), NeurIPS (2026), SPAA, ESA (2025), WebConf. (2024, 2025, 2026), ICLR (2023), SODA (2019, 2023)