Distributed Gaussian Mean Estimation under Communication Constraints: Optimal Rates and Communication-Efficient Algorithms
This work addresses the problem of efficient statistical estimation in distributed systems with limited communication, which is crucial for applications like federated learning and sensor networks, representing a foundational advancement rather than an incremental improvement.
The paper tackles distributed Gaussian mean estimation under communication constraints, establishing minimax rates of convergence and developing communication-efficient, statistically optimal procedures, with results showing that in the univariate case, the optimal rate depends only on the total communication budget, while in the multivariate case, it depends on specific budget allocations among machines.
We study distributed estimation of a Gaussian mean under communication constraints in a decision theoretical framework. Minimax rates of convergence, which characterize the tradeoff between the communication costs and statistical accuracy, are established in both the univariate and multivariate settings. Communication-efficient and statistically optimal procedures are developed. In the univariate case, the optimal rate depends only on the total communication budget, so long as each local machine has at least one bit. However, in the multivariate case, the minimax rate depends on the specific allocations of the communication budgets among the local machines. Although optimal estimation of a Gaussian mean is relatively simple in the conventional setting, it is quite involved under the communication constraints, both in terms of the optimal procedure design and lower bound argument. The techniques developed in this paper can be of independent interest. An essential step is the decomposition of the minimax estimation problem into two stages, localization and refinement. This critical decomposition provides a framework for both the lower bound analysis and optimal procedure design.