My research is at the intersection of machine learning and data privacy. My primary research interest is in designing privacy preserving machine learning algorithms with strong analytical guarantees, which are robust to errors in the data. In many instances I harness the privacy property of the algorithms to obtain robustness and utility guarantees. My combination of academic and industrial experience has allowed me to draw non-obvious insights at the intersection of theoretical analysis and practical deployment of privacy preserving machine learning algorithms. More specifically, my research aims to establish a strong coupling between differential privacy and machine learning by exploring the following two propositions:
-- Robust learning aids privacy: Any machine learning algorithm with good predictive ability (generalizability) can be made differentially private.
-- Privacy aids robust learning: Any differentially private machine learning algorithm, with low error rates in the training data, has good predictive ability.
