DeepAI AI Chat
Log In Sign Up

FetchSGD: Communication-Efficient Federated Learning with Sketching

by   Daniel Rothchild, et al.

Existing approaches to federated learning suffer from a communication bottleneck as well as convergence issues due to sparse client participation. In this paper we introduce a novel algorithm, called FetchSGD, to overcome these challenges. FetchSGD compresses model updates using a Count Sketch, and then takes advantage of the mergeability of sketches to combine model updates from many workers. A key insight in the design of FetchSGD is that, because the Count Sketch is linear, momentum and error accumulation can both be carried out within the sketch. This allows the algorithm to move momentum and error accumulation from clients to the central aggregator, overcoming the challenges of sparse client participation while still achieving high compression rates and good convergence. We prove that FetchSGD has favorable convergence guarantees, and we demonstrate its empirical effectiveness by training two residual networks and a transformer model.


page 1

page 2

page 3

page 4


Improved Convergence Rates for Non-Convex Federated Learning with Compression

Federated learning is a new distributed learning paradigm that enables e...

Comfetch: Federated Learning of Large Networks on Memory-Constrained Clients via Sketching

A popular application of federated learning is using many clients to tra...

Faster Rates for Compressed Federated Learning with Client-Variance Reduction

Due to the communication bottleneck in distributed and federated learnin...

SCAFFOLD: Stochastic Controlled Averaging for On-Device Federated Learning

Federated learning is a key scenario in modern large-scale machine learn...

Double Momentum SGD for Federated Learning

Communication efficiency is crucial in federated learning. Conducting ma...

On the Convergence of Momentum-Based Algorithms for Federated Stochastic Bilevel Optimization Problems

In this paper, we studied the federated stochastic bilevel optimization ...

Compressing Gradient Optimizers via Count-Sketches

Many popular first-order optimization methods (e.g., Momentum, AdaGrad, ...