Adaptive distributed methods under communication constraints
Statistics Theory
2019-02-06 v2 Machine Learning
Statistics Theory
Abstract
We study distributed estimation methods under communication constraints in a distributed version of the nonparametric random design regression model. We derive minimax lower bounds and exhibit methods that attain those bounds. Moreover, we show that adaptive estimation is possible in this setting.
Cite
@article{arxiv.1804.00864,
title = {Adaptive distributed methods under communication constraints},
author = {Botond Szabo and Harry van Zanten},
journal= {arXiv preprint arXiv:1804.00864},
year = {2019}
}
Comments
46 pages