Private and Communication-Efficient Algorithms for Entropy Estimation. (arXiv:2305.07751v1 [cs.LG])

Modern statistical estimation is often performed in a distributed setting
where each sample belongs to a single user who shares their data with a central
server. Users are typically concerned with preserving the privacy of their
samples, and also with minimizing the amount of data they must transmit to the
server. We give improved private and communication-efficient algorithms for
estimating several popular measures of the entropy of a distribution. All of
our algorithms have constant communication cost and satisfy local …

algorithms communication data distributed entropy privacy private server single