Skip to main content

Research Lectures

Leveraging Randomness in Structure to Enable Efficient Distributed Data Analytics – UCSD – October 2015

Preserving privacy in distributed analytics is important, but often it significantly increases the computational costs, to the point where such distributed learning becomes infeasible. In this talk, I presented an alternative approach where we can leverage the randomness in structure for a particular learning algorithm to also provide privacy while improving on the efficiency.