Det-CGD: Compressed Gradient Descent with Matrix Stepsizes for Non-Convex Optimization
Keywords: Non-Convex Optimization, First-Order Methods, Compression, Federated Learning
TL;DR: In this paper we propose a matrix stepsized CGD algorithm.
Abstract: This paper introduces a new method for minimizing matrix-smooth non-convex objectives through the use of novel Compressed Gradient Descent (CGD) algorithms enhanced with a matrix-valued stepsize. The proposed algorithms are theoretically analyzed first in the single-node and subsequently in the distributed settings. Our theoretical results reveal that the matrix stepsize in CGD can capture the objective’s structure and lead to faster convergence compared to a scalar stepsize. As a byproduct of our general results, we emphasize the importance of selecting the compression mechanism and the matrix stepsize in a layer-wise manner, taking advantage of model structure. Moreover, we provide theoretical guarantees for free compression, by designing specific layer-wise compressors for the non-convex matrix smooth objectives. Our findings are supported with empirical evidence.
Submission Number: 17
Loading