The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.
1-bit stochastic gradient descent and its application to data-parallel distributed training of speech DNNs
Abstract
We show empirically that in SGD training of deep neural networks, one can, at no or nearly no loss of accuracy, quantize the gradients aggressively—to but one bit per value—if the quantization error is carried forward across minibatches (error feedback). This size reduction makes it feasible to parallelize SGD through data-parallelism with fast processors like recent GPUs. We implement data-parallel deterministically distributed SGD by combining this finding with AdaGrad, automatic minibatch-size selection, double buffering, and model parallelism. Unexpectedly, quantization benefits AdaGrad, giving a small accuracy gain. For a typical Switchboard DNN with 46M parameters, we reach computation speeds of 27k frames per second (kfps) when using 2880 samples per minibatch, and 51kfps with 16k, on a server with 8 K20X GPUs. This corresponds to speed-ups over a single GPU of 3.6 and 6.3, respectively. 7 training passes over 309h of data complete in under 7h. A 160M-parameter model training processes 3300h of data in under 16h on 20 dual-GPU servers—a 10 times speed-up—albeit at a small accuracy loss.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Learning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892
Cited by
Sorted by citations
Convergence of Edge Computing and Deep Learning: A Comprehensive Survey
Wang, X., Han, Y., Leung, V. C. M., Niyato, D., Yan, X., & Chen, X. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2970550 · Cited by 1,498
“, data parallelism and model parallelism [80]–[83] as illustrated in Fig.”
MethodFederated learning on non-IID data: A survey
Zhu, H., Xu, J., Liu, S., & Jin, Y. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.07.098 · Cited by 1,062
FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantization
Reisizadeh, A., Mokhtari, A., Hassani, H., Jadbabaie, A., & Pedarsani, R. (2019) · arXiv · doi:10.48550/arxiv.1909.13014 · Cited by 587
Machine Learning at the Wireless Edge: Distributed Stochastic Gradient Descent Over-the-Air
Mohammadi Amiri, M., & Gunduz, D. (2020) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2020.2981904 · Cited by 548
The Microsoft 2017 Conversational Speech Recognition System
Xiong, W., Wu, L., Alleva, F., Droppo, J., Huang, X., & Stolcke, A. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8461870 · Cited by 472
Self-citationCommunication-efficient edge ai: algorithms and systems
Shi, Y., Yang, K., Jiang, T., Zhang, J., & Letaief, K. B. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.3007787 · Cited by 488
The microsoft 2016 conversational speech recognition system
Xiong, W., Droppo, J., Huang, X., Seide, F., Seltzer, M., Stolcke, A., Yu, D., & Zweig, G. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7953159 · Cited by 326
Self-citationQsparse-local-SGD: Distributed SGD with quantization, sparsification, and local computations
Basu, D., Data, D., Karakus, C., & Diggavi, S. N. (2020) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2020.2985917 · Cited by 256
Toward Human Parity in Conversational Speech Recognition
Xiong, W., Droppo, J., Huang, X., Seide, F., Seltzer, M. L., Stolcke, A., Yu, D., & Zweig, G. (2017) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2017.2756440 · Cited by 245
Self-citationSparse Binary Compression: Towards Distributed Deep Learning with minimal Communication
Sattler, F., Wiedemann, S., Muller, K.-R., & Samek, W. (2019) · 2019 International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2019.8852172 · Cited by 240
Privacy-Preserving Asynchronous Federated Learning Mechanism for Edge Network Computing
Lu, X., Liao, Y., Lio, P., & Hui, P. (2020) · IEEE Access · doi:10.1109/access.2020.2978082 · Cited by 159
Deep learning in the automotive industry: Applications and tools
Luckow, A., Cook, M., Ashcraft, N., Weill, E., Djerekarov, E., & Vorster, B. (2016) · 2016 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata.2016.7841045 · Cited by 164
PipeDream: Fast and Efficient Pipeline Parallel DNN Training
Harlap, A., Narayanan, D., Phanishayee, A., Seshadri, V., Devanur, N., Ganger, G., & Gibbons, P. (2018) · arXiv · doi:10.48550/arxiv.1806.03377 · Cited by 153
“CNTK’s [33] 1-bit quantization technique tackles the problem of communication bottlenecks in data parallelism training [35].”
MethodLi, A., Sun, J., Zeng, X., Zhang, M., Li, H., & Chen, Y. (2021) · Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems · doi:10.1145/3485730.3485929 · Cited by 139
Qian, B., Su, J., Wen, Z., Jha, D. N., Li, Y., Guan, Y., Puthal, D., James, P., Yang, R., Zomaya, A. Y., et al. (2021) · ACM Computing Surveys · doi:10.1145/3398020 · Cited by 132
A Review of Privacy-Preserving Federated Learning for the Internet-of-Things
Briggs, C., Fan, Z., & Andras, P. (2021) · Studies in Computational Intelligence · doi:10.1007/978-3-030-70604-3_2 · Cited by 101
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
Kerkouche, R., Ács, G., Castelluccia, C., & Genevès, P. (2021) · Proceedings of the Conference on Health, Inference, and Learning · doi:10.1145/3450439.3451859 · Cited by 65
Taming unbalanced training workloads in deep learning with partial collective operations
Li, S., Ben-Nun, T., Girolamo, S. D., Alistarh, D., & Hoefler, T. (2020) · Proceedings of the 25th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming · doi:10.1145/3332466.3374528 · Cited by 57
rTop-k: A Statistical Estimation Approach to Distributed SGD
Barnes, L. P., Inan, H. A., Isik, B., & Ozgur, A. (2020) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2020.3042094 · Cited by 44
Metadata
- Authors
- Seide, F., Fu, H., Droppo, J., Li, G., & Yu, D.
- Venue
- Interspeech 2014
- Publisher
- ISCA
- Published
- 2014-09-14
- Pages
- 1058-1062
- Cited by
- 443
- References
- 1
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
