picture of me taken by Sury.

Daniel Alabi

alabid [at] g [dot] harvard [dot] edu
Harvard University
Advised by Salil Vadhan
Ph.D. student in the Theory of Computation group and the Harvard Privacy Tools group
Photo credit: My neighbor Sury @scrodcity

Currently, my primary research interest is in the Algorithmic Foundations of Data Privacy: the study of the design, analysis, and limitations of algorithms in settings where we need to ensure individual data privacy guarantees.

In particular, I'm interested in the tradeoffs in computational and statistical resources that result when we require individual privacy guarantees and, as a consequence, the robustness of statistical models and machine learning predictors. Example computational resources include time, memory, randomness, communication, and parallelism. Example statistical resources include samples (or labeled examples for classification or regression) drawn from an unknown distribution.

More generally, my research interests lie in theoretical computer science (algorithms and complexity theory).

Funding Acknowledgements: My research was/is supported by Harvard SEAS, Harvard CRCS, the Facebook Emerging Scholar Award, and the Courtlandt S. Gross Memorial Scholarship.

Personal: I am Nigerian and a member of Harvard's competitive Ballroom Dance Team.

Selected Research

The Cost of a Reductions Approach to Private Fair Optimization
Full version in submission  [arxiv] 
To appear in TPDP 2019 in November 2019
Poster at the Recent Themes in Resource Tradeoffs: Privacy, Fairness, and Robustness workshop at the University of Minnesota in June 2019 
Learning to Prune: Speeding up Repeated Computations
with Adam Tauman Kalai, Katrina Ligett, Cameron Musco, Christos Tzamos, and Ellen Vitercik
COLT, June, 2019  [arxiv]  [PMLR] 
Unleashing Linear Optimizers for Group-Fair Learning and Optimization
with Nicole Immorlica and Adam Tauman Kalai
COLT, July, 2018  [arxiv]  [PMLR] 
Learning Certifiably Optimal Rule Lists for Categorical Data
with Elaine Angelino, Nicholas Larus-Stone, Margo Seltzer, and Cynthia Rudin
JMLR, June, 2018  [arxiv]  [JMLR] 

See all papers here.

Selected Code Artifacts

Collaborative LaTeX Editor
CORELS is a custom discrete optimization technique for building rule lists over a categorical feature space.
See my GitHub page for some more code artifacts not listed here.


Check out some of my instructional worksheets.