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 accessAbstract
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.
Integrity signals
Registry
Cites
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.
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?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Cross-device federated learning and federated data analysis are now being applied in consumer digital products.”
From Advances and Open Problems in Federated Learning (2021) · cites this paper as [112]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Google makes extensive use of federated learning in the Gboard mobile keyboard [376, 222, 491, 112, 383], as well as in features on Pixel phones [14] and in Android Messages [439].”
From Advances and Open Problems in Federated Learning (2021) · cites this paper as [112]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“When local datasets are small and the data is IID, FL clearly has an edge, and indeed, real-world applications of federated learning [491, 222, 112] benefit from training a single model across devices.”
From Advances and Open Problems in Federated Learning (2021) · cites this paper as [112]
Experimental — accuracy not yet measured
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
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Boyd, S., & Vandenberghe, L. (2004) · doi:10.1017/cbo9780511804441 · Cited by 56,601
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
Communication-Efficient Learning of Deep Networks from Decentralized Data
McMahan, H. B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. Y. (2016) · arXiv · doi:10.48550/arxiv.1602.05629 · Cited by 14,215
A Stochastic Approximation Method
Robbins, H., & Monro, S. (1951) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177729586 · Cited by 9,643
(2010) · Cited by 8,261
Federated learning: Challenges, methods, and future directions
Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020) · IEEE Signal Processing Magazine · doi:10.1109/msp.2020.2975749 · Cited by 6,210
An overview of gradient descent optimization algorithms
Ruder, S. (2016) · arXiv · doi:10.48550/arxiv.1609.04747 · Cited by 4,962
Federated Learning: Strategies for Improving Communication Efficiency
Konečný, J., McMahan, H. B., Yu, F. X., Richtárik, P., Suresh, A. T., & Bacon, D. (2016) · arXiv · doi:10.48550/arxiv.1610.05492 · Cited by 3,766
Large Scale Distributed Deep Networks
Dean, J. B., Corrado, G. S., Monga, R., Chen, K., Devin, M., Mao, M., Ranzato, M., Senior, A., Tucker, P. A., Yang, K., et al. (2012) · Cited by 3,199
Research works printed but not linked: 2
- al-lawati, 2019, Anytime minibatch with stale gradients, Proc 53rd Annu Conf Inf Sci Syst (CISS), 1
- cadambe, 2017, Codes for distributed computing: A tutorial, IEEE Inf Theory Soc Newslett, 67, 3
Cited by
Advances and Open Problems in Federated Learning
Kairouz, P., & McMahan, H. B. (2021) · Foundations and Trends in Machine Learning · doi:10.1561/2200000083 · Cited by 7,772
Linear Convergence in Federated Learning: Tackling Client Heterogeneity and Sparse GradientsPreprint
Mitra, A., Jaafar, R. H., Pappas, G. J., & Hassani, H. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.07053 · Cited by 203
Multi-Stage Asynchronous Federated Learning With Adaptive Differential Privacy
Li, Y., Yang, S.-S., Ren, X.-B., Shi, L., & Zhao, C. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3332428 · Cited by 47
Stigmergic Independent Reinforcement Learning for Multiagent Collaboration
Xu, X., Li, R., Zhao, Z., & Zhang, H. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3056418 · Cited by 42
Robust Federated Learning for Unreliable and Resource-Limited Wireless Networks
Chen, Z., Yi, W., Liu, Y., & Nallanathan, A. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3366393 · Cited by 50
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
- green
- arxiv.org
- Licence
- 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.