Sotabase

Career

· Thomas D. Cabot Associate Professor of Electrical Engineering, Harvard University2024–
· Assistant Professor of Electrical Engineering, Harvard University2017–2024
· Inaugural Data Science for Social Good Post-Doctoral Fellow, IBM Research2015–2017
· Ph.D. in Electrical Engineering and Computer Science, Massachusetts Institute of Technology (MIT)2009–2015
· B.E. in Communications Engineering, Universidade de Brasilia
· M.Sc. in Electrical Engineering, Universidade Estadual de Campinas (Unicamp)

Publications (139)

Optimized Pre-Processing for Discrimination Prevention
Neural Information Processing Systems · 2017
814
cited
Privacy against statistical inference
Allerton Conference on Communication, Control, and Computing · 2012
361
cited
Predictive Multiplicity in Classification
International Conference on Machine Learning · 2019
182
cited
Repairing without Retraining: Avoiding Disparate Impact with Counterfactual Distributions
International Conference on Machine Learning · 2019
94
cited
CPR: Classifier-Projection Regularization for Continual Learning
International Conference on Learning Representations · 2020
90
cited
Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE)
Neural Information Processing Systems · 2024
84
cited
Principal Inertia Components and Applications
IEEE Transactions on Information Theory · 2017
82
cited
Fundamental limits of perfect privacy
International Symposium on Information Theory · 2015
80
cited
Managing Your Private and Public Data: Bringing Down Inference Attacks Against Your Privacy
IEEE Journal on Selected Topics in Signal Processing · 2014
76
cited
How to hide the elephant- or the donkey- in the room: Practical privacy against statistical inference for large data
IEEE Global Conference on Signal and Information Processing · 2013
68
cited
Hypothesis testing under maximal leakage privacy constraints
International Symposium on Information Theory · 2017
68
cited
An Information-Theoretic View of Generalization via Wasserstein Distance
International Symposium on Information Theory · 2019
66
cited
Optimized Score Transformation for Fair Classification
International Conference on Artificial Intelligence and Statistics · 2019
64
cited
Optimized Data Pre-Processing for Discrimination Prevention
arXiv.org · 2017
62
cited
Multi-Path TCP with Network Coding for Mobile Devices in Heterogeneous Networks
IEEE Vehicular Technology Conference · 2013
54
cited
On the Robustness of Information-Theoretic Privacy Measures and Mechanisms
IEEE Transactions on Information Theory · 2018
54
cited
A Tunable Measure for Information Leakage
International Symposium on Information Theory · 2018
52
cited
Beyond Adult and COMPAS: Fair Multi-Class Prediction via Information Projection
Neural Information Processing Systems · 2022
52
cited
Rashomon Capacity: A Metric for Predictive Multiplicity in Classification
Neural Information Processing Systems · 2022
52
cited
Privacy With Estimation Guarantees
IEEE Transactions on Information Theory · 2017
51
cited
Sotabase