Private PAC learning implies finite Littlestone dimension
Proceedings of the 51st Annual ACM SIGACT Symposium on Theory of Computing, Volume abs/1806.00949, 2019, Pages 852-860.
We show that every approximately differentially private learning algorithm (possibly improper) for a class H with Littlestone dimension d requires Ω(log*(d)) examples. As a corollary it follows that the class of thresholds over ℕ can not be learned in a private manner; this resolves open questions due to [Bun et al. 2015] and [Feldman and...More
PPT (Upload PPT)