Adaptive distributed methods under communication constraints
arXiv:1804.00864v229 citations
AI Analysis
This work addresses communication-efficient distributed learning for statistical estimation, representing an incremental advance in optimizing performance under constraints.
The paper tackles the problem of distributed estimation under communication constraints in nonparametric random design regression, deriving minimax lower bounds and presenting methods that achieve these bounds while enabling adaptive estimation.
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.