Sparsification as a Remedy for Staleness in Distributed Asynchronous SGD - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year :

Sparsification as a Remedy for Staleness in Distributed Asynchronous SGD

Maurizio Filippone
  • Function : Author
  • PersonId : 1021042
Pietro Michiardi
  • Function : Author
  • PersonId : 1084771

Abstract

Large scale machine learning is increasingly relying on distributed optimization, whereby several machines contribute to the training process of a statistical model. In this work we study the performance of asynchronous, distributed settings, when applying sparsification, a technique used to reduce communication overheads. In particular, for the first time in an asynchronous, non-convex setting, we theoretically prove that, in presence of staleness, sparsification does not harm SGD performance: the ergodic convergence rate matches the known result of standard SGD, that is $\mathcal{O} \left( 1/\sqrt{T} \right)$. We also carry out an empirical study to complement our theory, and confirm that the effects of sparsification on the convergence rate are negligible, when compared to 'vanilla' SGD, even in the challenging scenario of an asynchronous, distributed system.

Dates and versions

hal-03345253 , version 1 (15-09-2021)

Identifiers

Cite

Rosa Candela, Giulio Franzese, Maurizio Filippone, Pietro Michiardi. Sparsification as a Remedy for Staleness in Distributed Asynchronous SGD. 2019. ⟨hal-03345253⟩
11 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More