Privacy-preserving Principal Component Analysis: Results in Progress and Results in Perspectives

16.12.2015 10:15


Infotech Oulu Lecture Series

Lecturer: Dr. Lu Wei, Harvard University. USA

Date: December 16, 2015
Time: 10:15-12:00
Room: TS127


In a recent work, Chaudhuri, Sarwate, and Sinha studied the performance of a privacy-preserving PCA algorithm in the presence of one principal component. We extend the analysis to an arbitrary number of principal components via tools from random matrix theory. Some related open problems will also be discussed. (Joint work with Tarokh, Sarwate, and Corander).

Short bio

Lu Wei obtained his doctoral degree (with distinction) from Aalto University, Finland, in 2013. He is currently a Postdoctoral Fellow jointly appointed by Harvard University and University of Helsinki. His
research interests lie in random matrix theory with applications to wireless communications, machine learning, and data sciences.

More information: Markku Juntti

Last updated: 9.12.2015