DRIVE: One-bit Distributed Mean Estimation

Shay Vargaftik, Ran Ben-Basat, Amit Portnoy, Gal Mendelson, Yaniv Ben-Itzhak, Michael Mitzenmacher

Keywords: Distributed Mean Estimation, Stochastic Approximation, Federated Learning, Distributed Leaning, Communication Efficiency, Bandwidth Reduction

TL;DR: Novel mathematical results and new algorithms for distributed mean estimation using one bit per coordinate.

Abstract: We consider the problem where $n$ clients transmit $d$-dimensional real-valued vectors using $d(1+o(1))$ bits each, in a manner that allows the receiver to approximately reconstruct their mean. Such compression problems naturally arise in distributed and federated learning. We provide novel mathematical results and derive computationally efficient algorithms that are more accurate than previous compression techniques. We evaluate our methods on a collection of distributed and federated learning tasks, using a variety of datasets, and show a consistent improvement over the state of the art.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/amitport/DRIVE-One-bit-Distributed-Mean-Estimation

Community Implementations: [ 5 code implementations](https://www.catalyzex.com/paper/arxiv:2105.08339/code)