Lazily Aggregated Quantized Gradient Innovation for Communication-Efficient Federated Learning
The DOI matches its Crossref record, with a small difference in title or year; 38 of the 39 references it cites were found in a registry or an index.
Abstract
This paper focuses on communication-efficient federated learning problem, and develops a novel distributed quantized gradient approach, which is characterized by adaptive communications of the quantized gradients. Specifically, the federated learning builds upon the server-worker infrastructure, where the workers calculate local gradients and upload them to the server; then the server obtain the global gradient by aggregating all the local gradients and utilizes it to update the model parameter. The key idea to save communications from the worker to the server is to quantize gradients as well as skip less informative quantized gradient communications by reusing previous gradients. Quantizing and skipping result in 'lazy' worker-server communications, which justifies the term Lazily Aggregated Quantized (LAQ) gradient. Theoretically, the LAQ algorithm achieves the same linear convergence as the gradient descent in the strongly convex case, while effecting major savings in the communication in terms of transmitted bits and communication rounds. Empirically, extensive experiments using realistic data corroborate a significant communication reduction compared with state-of-the-art gradient- and stochastic gradient-based algorithms.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“In particular, a lazily aggregated quantized gradient method was further proposed in [132] to reduce both the amount of exchanged data and communication rounds by reusing the outdated gradients for the less informative quantized gradients.”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [132]
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
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
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
Zhu, L., & Han, S. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-63076-8_2 · Cited by 1,906
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 · cited 2×
“To reduce information losses, gradient components with small values are accumulated and then applied when they exceed a certain threshold [18].”
BackgroundMethodQSGD: Communication-Efficient SGD via Gradient Quantization and Encoding
Alistarh, D., Grubic, D., Li, J., Tomioka, R., & Vojnović, M. (2016) · London School of Economics and Political Science Research Online (London School of Economics and Political Science) · doi:10.48550/arxiv.1610.02132 · Cited by 859
Communication Efficient Distributed Machine Learning with the Parameter Server
Li, M., Andersen, D. G., Smola, A. J., & Yu, K. (2014) · neural information processing systems · Cited by 715
Sparse Communication for Distributed Gradient Descent
Aji, A. F., & Heafield, K. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1045 · Cited by 585
Deep learning with Elastic Averaging SGDPreprint
Zhang, S., Choromanska, A., & LeCun, Y. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1412.6651 · Cited by 546
signSGD: Compressed Optimisation for Non-Convex Problems
Bernstein, J., Wang, Y.-X., Azizzadenesheli, K., & Anandkumar, A. (2018) · arXiv · doi:10.48550/arxiv.1802.04434 · Cited by 541
The Convergence of Sparsified Gradient MethodsPreprint
Alistarh, D., Hoefler, T., Johansson, M., Khirirat, S., Konstantinov, N., & Renggli, C. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.10505 · Cited by 525
Scalable distributed DNN training using commodity GPU cloud computing
Strom, N. (2015) · Interspeech 2015 · doi:10.21437/interspeech.2015-354 · Cited by 521
Stich, S. U., Cordonnier, J.-B., & Jaggi, M. (2018) · arXiv · doi:10.48550/arxiv.1809.07599 · Cited by 508
Seide, F., Fu, H., Droppo, J., Li, G., & Yu, D. (2014) · Interspeech 2014 · doi:10.21437/interspeech.2014-274 · Cited by 443
Communication-Efficient Distributed Statistical Inference
Jordan, M. I., Lee, J. D., & Yang, Y. (2019) · Journal of the American Statistical Association · doi:10.1080/01621459.2018.1429274 · Cited by 442
Computer networks: A systems approach
Parker, B. H. (1997) · Journal of the American Society for Information Science · doi:10.1002/(sici)1097-4571(199709)48:9<861::aid-asi11>3.0.co;2-u · Cited by 434
Communication-Efficient Distributed Optimization using an Approximate Newton-type Method
Shamir, O., Srebro, N., & Zhang, T. (2013) · arXiv · doi:10.48550/arxiv.1312.7853 · Cited by 424 · cited 1×
“Apart from the aforementioned gradient compression approaches, communication-efficient schemes aiming to reduce communication rounds have been developed by leveraging higher-order information [26, 27], periodic aggregation [4, 28, 29], and recently by adaptive aggregation [30, 31, 32]; see also [33] for a lower bound…”
BackgroundLAG: Lazily Aggregated Gradient for Communication-Efficient Distributed LearningPreprint
Chen, T., Giannakis, G. B., Sun, T., & Yin, W. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.09965 · Cited by 329
Self-citationDiSCO: Distributed Optimization for Self-Concordant Empirical Loss
Zhang, Y., & Xiao, L. (2015) · International Conference on Machine Learning · Cited by 243
A linear speedup analysis of distributed deep learning with sparse and quantized communication
Jiang, P., & Agrawal, G. (2018) · neural information processing systems · Cited by 218
ZipML: Training Linear Models with End-to-End Low Precision, and a Little Bit of Deep Learning
Zhang, H., Li, J. Z., Kara, K., Alistarh, D., Liu, J., & Zhang, C. (2017) · International Conference on Machine Learning · Cited by 206
Gradient Sparsification for Communication-Efficient Distributed Optimization
Wangni, J., Wang, J., Ji, L. H., & Zhang, T. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1710.09854 · Cited by 198
Error Compensated Quantized SGD and its Applications to Large-scale Distributed OptimizationPreprint
Wu, J., Huang, W., Huang, J., & Zhang, T. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.08054 · Cited by 193
ATOMO: Communication-efficient Learning via Atomic Sparsification
Wang, H., Sievert, S., Charles, Z., Liu, S., Wright, S. J., & Papailiopoulos, D. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.04090 · Cited by 189
Sensor-Centric Data Reduction for Estimation With WSNs via Censoring and Quantization
Msechu, E. J., & Giannakis, G. B. (2011) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2011.2171686 · Cited by 156
Self-citationAsynchronous Stochastic Gradient Descent with Delay Compensation
Zheng, S., Meng, Q., Wang, T., Chen, W., Yu, N., Ma, Z.-M., & Liu, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1609.08326 · Cited by 154
Wang, J., & Joshi, G. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.07576 · Cited by 146
Error Feedback Fixes SignSGD and other Gradient Compression Schemes
Karimireddy, S. P., Rebjock, Q., Stich, S. U., & Jaggi, M. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.09847 · Cited by 137
Communication Complexity of Distributed Convex Learning and Optimization
Arjevani, Y., & Shamir, O. (2015) · arXiv · doi:10.48550/arxiv.1506.01900 · Cited by 133 · cited 1×
“…communication-efficient schemes aiming to reduce communication rounds have been developed by leveraging higher-order information [26, 27], periodic aggregation [4, 28, 29], and recently by adaptive aggregation [30, 31, 32]; see also [33] for a lower bound on the number of communication rounds.”
BackgroundOn the Convergence Rate of Incremental Aggregated Gradient Algorithms
Gürbüzbalaban, M., Ozdaglar, A., & Parrilo, P. A. (2017) · SIAM Journal on Optimization · doi:10.1137/15m1049695 · Cited by 121
Distributed Learning with Compressed Gradient Differences
Mishchenko, K., Gorbunov, E., Takáč, M., & Richtárik, P. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.09269 · Cited by 97
Randomized distributed mean estimation: Accuracy vs. communication
Konečný, J., & Richtárik, P. (2018) · Frontiers in Applied Mathematics and Statistics · doi:10.3389/fams.2018.00062 · Cited by 69 · cited 1×
“Quantization and sparsification can also be employed jointly [3, 23, 24].”
BackgroundCommunication-Efficient Distributed Learning via Lazily Aggregated Quantized Gradients
Sun, J., Chen, T., Giannakis, G. B., & Yang, Z. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.07588 · Cited by 56
Self-citationGradient Perturbation is Underrated for Differentially Private Convex Optimization
Yu, D., Zhang, H., Chen, W., Yin, J., & Liu, T. (2020) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/431 · Cited by 48
Yu, H., & Jin, R. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.04346 · Cited by 45
Communication-Efficient Distributed Reinforcement LearningPreprint
Chen, T., Zhang, K., Giannakis, G. B., & Başar, T. (2018) · arXiv (Cornell University) · Cited by 35
Self-citationMagnússon, S., Shokri‐Ghadikolaei, H., & Li, N. (2019) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/ieeeconf44664.2019.9049052 · Cited by 25
Research works printed but not linked: 3
- McMahan, Communication-efficient learning of deep networks from decentralized data, 1273
- 10.5555/3294771.3294915 · doi:10.5555/3294771.3294915
- LeCun, 2010, MNIST handwritten digit database
Cited by
Edge artificial intelligence for 6g: vision, enabling technologies, and applications
Letaief, K. B., Shi, Y., Lu, J., & Lu, J. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3126076 · Cited by 829
Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmark
Huang, W., Ye, M., Shi, Z., Wan, G., Li, H., Du, B., & Yang, Q. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3418862 · Cited by 232
Self-citationMoshawrab, M., Adda, M., Bouzouane, A., Ibrahim, H., & Raad, A. (2023) · Electronics · doi:10.3390/electronics12102287 · Cited by 219
EF21: A New, Simpler, Theoretically Better, and Practically Faster Error FeedbackPreprint
Richtárik, P., Sokolov, I., & Fatkhullin, I. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.05203 · Cited by 196
Communication-Efficient Federated Learning with Adaptive Quantization
Mao, Y., Zhao, Z., Yan, G., Liu, Y., Lan, T., Song, L., & Ding, W. (2022) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3510587 · Cited by 126
Metadata
- Authors
- Sun, J., Chen, T., Giannakis, G. B., Yang, Q., & Yang, Z.
- Affiliations
- State Key Laboratory of Industrial Control Technology (CN)
- Rensselaer Polytechnic Institute (US)
- University of Minnesota (US)
- Southern University of Science and Technology (CN)
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2022-04-01
- Volume
- 44
- Issue
- 4
- Pages
- 2031-2044
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- 1939-3539
- OpenAlex
- W3093828967
- Cited by
- 102
- References
- 39
- Citations by year
- 2026: 10 · 2025: 16 · 2024: 28 · 2023: 26 · 2022: 13
- FWCI
- 4.71
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- Funders
- National Natural Science Foundation of China (61873118, 61751205)
- Shenzhen Committee on Science and Innovations (GJHZ20180411143603361)
- Guangdong Science and Technology Department (2018A050506003)
- China Scholarship Council
- Key-Area Research and Development Program of Guangdong Province (2018B010107002)
- National Science Foundation (1901134)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Physical Sciences › Engineering › Control and Systems Engineering › Distributed Control Multi-Agent Systems
- Keywords
- federated learning, communication efficiency, gradient quantization, adaptive communication, linear convergence, strongly convex optimization, distributed optimization
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
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.