Asynchrony begets momentum, with an application to deep learning
Open accessThe DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.
Abstract
Asynchronous methods are widely used in deep learning, but have limited theoretical justification when applied to non-convex problems. We show that running stochastic gradient descent (SGD) in an asynchronous manner can be viewed as adding a momentum-like term to the SGD iteration. Our result does not assume convexity of the objective function, so it is applicable to deep learning systems. We observe that a standard queuing model of asynchrony results in a form of momentum that is commonly used by deep learning practitioners. This forges a link between queuing theory and asynchrony in deep learning systems, which could be useful for systems builders. For convolutional neural networks, we experimentally validate that the degree of asynchrony directly correlates with the momentum, confirming our main result. An important implication is that tuning the momentum parameter is important when considering different levels of asynchrony. Furthermore, our theory suggests ways of counter-acting the adverse effects of asynchrony. We see that a simple mechanism like using negative algorithmic momentum can be beneficial under high asynchrony. Since asynchronous methods have better hardware efficiency, this result may shed light on when asynchronous execution is more efficient for deep learning systems.
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.
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
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Large-Scale Machine Learning with Stochastic Gradient Descent
Bottou, L. (2010) · Proceedings of COMPSTAT'2010 · doi:10.1007/978-3-7908-2604-3_16 · Cited by 5,795 · cited 1×
“(1) Typically the component function fi(w) represents a loss function evaluated on a specific data point zi.”
BackgroundSome methods of speeding up the convergence of iteration methods
Polyak, B. T. (1964) · USSR Computational Mathematics and Mathematical Physics · doi:10.1016/0041-5553(64)90137-5 · Cited by 3,069
Hogwild: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
Niu, F., Recht, B., Re, C., & Wright, S. J. (2011) · arXiv · doi:10.48550/arxiv.1106.5730 · Cited by 1,211 · cited 3×
“Let v t denote the value read by the worker in charge of update t .”
BackgroundInfluential
Cited by
Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
Lin, Y., Han, S., Mao, H., Wang, Y., & Dally, W. J. (2017) · arXiv · doi:10.48550/arxiv.1712.01887 · Cited by 1,134
“Mitliagkas et al. (2016) discussed the staleness caused by asynchrony and attributed it to a term described as implicit momentum .”
BackgroundMethodInfluentialYou, Y., Zhang, Z., Hsieh, C.-J., Demmel, J., & Keutzer, K. (2018) · Proceedings of the 47th International Conference on Parallel Processing · doi:10.1145/3225058.3225069 · Cited by 384
Toward Communication-Efficient Federated Learning in the Internet of Things With Edge Computing
Sun, H., Li, S., Yu, F. R., Qi, Q., Wang, J., & Liao, J. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2994596 · Cited by 95
“Mitliagkas et al. [54] suggested that stale update component in asynchronous parallelism could be viewed as implicit momentum.”
Slow and Stale Gradients Can Win the Race
Dutta, S., Wang, J., & Joshi, G. (2021) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2021.3103770 · Cited by 40
Metadata
- Authors
- Mitliagkas, I., Zhang, C., Hadjis, S., & Re, C.
- Venue
- 2016 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton)
- Publisher
- IEEE
- Published
- 2016-09
- Pages
- 997-1004
- OpenAlex
- W2410569288
- Cited by
- 113
- References
- 4
- FWCI
- 9.26
- Type
- conference paper
- Language
- en
- Open access
- green
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: Asynchrony begets momentum, with an application to deep learning. https://app.cochat.ai/papers/10.1109/allerton.2016.7852343
- The paper
- Mitliagkas, I., Zhang, C., Hadjis, S., & Re, C. (2016). Asynchrony begets momentum, with an application to deep learning. 2016 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton). https://doi.org/10.1109/allerton.2016.7852343
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.