Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 90 of the 90 references it cites were found in a registry or an index.

Slow and Stale Gradients Can Win the Race

Open access
Dutta, Sanghamitra, Wang, Jianyu, Joshi, Gauri · 2021 · IEEE Journal on Selected Areas in Information Theory

Abstract

Distributed Stochastic Gradient Descent (SGD) when run in a synchronous manner, suffers from delays inruntimeas it waits for the slowest workers (stragglers). Asynchronous methods can alleviate stragglers, but cause gradient staleness that can adversely affect the convergenceerror. In this work, we present a novel theoretical characterization of the speedup offered by asynchronous methods by analyzing the trade-off between theerrorin the trained model and the actual trainingruntime(wallclock time). The main novelty in our work is that our runtime analysis considers random straggling delays, which helps us design and compare distributed SGD algorithms that strike a balance between straggling and staleness. We also provide a new error convergence analysis of asynchronous SGD variants without bounded or exponential delay assumptions. Finally, based on our theoretical characterization of the error-runtime trade-off, we propose a method of gradually varying synchronicity in distributed SGD and demonstrate its performance on the CIFAR10 dataset.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/jsait.2021.3103770

Cites

90 of 90 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 88 references

Research works printed but not linked: 2

  1. al-lawati, 2019, Anytime minibatch with stale gradients, Proc 53rd Annu Conf Inf Sci Syst (CISS), 1
  2. cadambe, 2017, Codes for distributed computing: A tutorial, IEEE Inf Theory Soc Newslett, 67, 3

Cited by

See all 40

Metadata

Authors
Dutta, S., Wang, J., & Joshi, G.
Affiliations
Carnegie Mellon University (US)
Venue
IEEE Journal on Selected Areas in Information Theory
Published
2021-09
Volume
2
Issue
3
Pages
1012-1024
ISSN
2641-8770 (ISSN-L)
OpenAlex
W3013308737
Cited by
40
References
88
Citations by year
2026: 4 · 2025: 3 · 2024: 14 · 2023: 5 · 2022: 5 · 2021: 8
FWCI
2.80
Type
journal article
Language
en
Open access
Locations
  • IEEE Journal on Selected Areas in Information Theory (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Funders
  • IBM Faculty Award
  • National Science Foundation CRII Award (CCF-1717314)
  • Qualcomm Innovation Fellowship
  • AISTATS 2018
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
  • Physical Sciences › Engineering › Control and Systems Engineering › Distributed Control Multi-Agent Systems
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
Keywords
distributed stochastic gradient descent, asynchronous SGD, convergence analysis, random delay, distributed training

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Slow and Stale Gradients Can Win the Race. https://app.cochat.ai/papers/10.1109/jsait.2021.3103770
The paper
Dutta, S., Wang, J., & Joshi, G. (2021). Slow and Stale Gradients Can Win the Race. IEEE Journal on Selected Areas in Information Theory. https://doi.org/10.1109/jsait.2021.3103770

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.