Winning Privately: The Differentially Private Lottery Ticket Mechanism

Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho

TL;DR: An end-to-end differentially private extension of the lottery ticket mechanism

Abstract: We propose the differentially private lottery ticket mechanism (DPLTM). An end-to-end differentially private training paradigm based on the lottery ticket hypothesis. Using ``high-quality winners", selected via our custom score function, DPLTM significantly outperforms state-of-the-art. We show that DPLTM converges faster, allowing for early stopping with reduced privacy budget consumption. We further show that the tickets from DPLTM are transferable across datasets, domains, and architectures. Our extensive evaluation on several public datasets provides evidence to our claims.

Keywords: Differentially private neural networks, lottery ticket hypothesis, differential privacy