Abstract
Distributed optimization methods are actively researched by optimization community. Due to applications in distributed machine learning, modern research directions include stochastic objectives, reducing communication frequency and time-varying communication network topology. Recently, an analysis unifying several centralized and decentralized approaches to stochastic distributed optimization was developed in Koloskova et al. (2020). In this work, we employ a Catalyst framework and accelerate the rates of Koloskova et al. (2020) in the case of low stochastic noise.
The work of E. Trimbach and A. Rogozin was supported by Andrei M. Raigorodskii Scholarship in Optimization. The research of A. Rogozin is supported by the Ministry of Science and Higher Education of the Russian Federation (Goszadaniye) No-075-00337-20-03, project No. 0714-2020-0005. This work started during Summer school at Sirius Institute.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Assran, M., Loizou, N., Ballas, N., Rabbat, M.: Stochastic gradient push for distributed deep learning. In: International Conference on Machine Learning, pp. 344–353. PMLR (2019)
Dvinskikh, D., Gasnikov, A.: Decentralized and parallelized primal and dual accelerated methods for stochastic convex programming problems. arXiv preprint arXiv:1904.09015 (2019)
Koloskova, A., Lin, T., Stich, S.U., Jaggi, M.: Decentralized deep learning with arbitrary communication compression. In: ICLR 2020 Conference Blind Submission (2020)
Koloskova, A., Loizou, N., Boreiri, S., Jaggi, M., Stich, S.U.: A unified theory of decentralized sgd with changing topology and local updates. In: International Conference on Machine Learning (2020)
Konecnỳ, J., McMahan, H.B., Ramage, D., Richtárik, P.: Federated optimization: Distributed machine learning for on-device intelligence. arXiv preprint arXiv:1610.02527 (2016)
Kovalev, D., Salim, A., Richtárik, P.: Optimal and practical algorithms for smooth and strongly convex decentralized optimization. arXiv preprint arXiv:2006.11773 (2020)
Kulunchakov, A., Mairal, J.: A generic acceleration framework for stochastic composite optimization. In: Advanced Neural Information Processing System (NeurIPS 2019), vol. 32 (2019)
Li, H., Fang, C., Yin, W., Lin, Z.: A sharp convergence rate analysis for distributed accelerated gradient methods. arXiv:1810.01053 (2018)
Li, H., Lin, Z.: Revisiting extra for smooth distributed optimization. arXiv preprint arXiv:2002.10110 (2020)
Li, Z., Shi, W., Yan, M.: A decentralized proximal-gradient method with network independent step-sizes and separated convergence rates. IEEE Trans. Sign. Process. 67(17), 4494–4506 (2019)
Lian, X., Zhang, C., Zhang, H., Hsieh, C.J., Zhang, W., Liu, J.: Can decentralized algorithms outperform centralized algorithms? a case study for decentralized parallel stochastic gradient descent. arXiv preprint arXiv:1705.09056 (2017)
Lin H., Mairal J., H.Z.: A universal catalyst for first-order optimization. arXiv preprint arXiv:1506.02186 (2015)
Lin, H., Mairal, J., Harchaoui, Z.: Catalyst acceleration for first-order convex optimization: from theory to practice. J. Mach. Learn. Res. 18(1), 7854–7907 (2018)
McMahan, B., Moore, E., Ramage, D., Hampson, S., y Arcas, B.A.: Communication-efficient learning of deep networks from decentralized data. In: Artificial Intelligence and Statistics, pp. 1273–1282. PMLR (2017)
McMahan, H.B., Moore, E., Ramage, D., y Arcas, B.A.: Federated learning of deep networks using model averaging. arXiv preprint arXiv:1602.05629 (2016)
Nedić, A., Olshevsky, A., Shi, W.: Achieving geometric convergence for distributed optimization over time-varying graphs. SIAM J. Optim. 27(4), 2597–2633 (2017)
Nedic, A., Ozdaglar, A.: Distributed subgradient methods for multi-agent optimization. IEEE Trans. Autom. Cont. 54(1), 48–61 (2009)
Pu, S., Shi, W., Xu, J., Nedich, A.: A push-pull gradient method for distributed optimization in networks. In: 2018 IEEE Conference Decision Control (CDC), pp. 3385–3390 (2018)
Qu, G., Li, N.: Accelerated distributed nesterov gradient descent. In: 2016 54th Annual Allerton Conference on Communication, Control, and Computing (2016)
Scaman, K., Bach, F., Bubeck, S., Lee, Y.T., Massoulié, L.: Optimal algorithms for smooth and strongly convex distributed optimization in networks. In: International Conference on Machine Learning, pp. 3027–3036 (2017)
Shi, W., Ling, Q., Wu, G., Yin, W.: Extra: an exact first-order algorithm for decentralized consensus optimization. SIAM J. Optim. 25(2), 944–966 (2015)
Stich, S.U.: Local sgd converges fast and communicates little. arXiv preprint arXiv:1805.09767 (2018)
Tang, H., Lian, X., Yan, M., Zhang, C., Liu, J.: D2: decentralized training over decentralized data. In: International Conference on Machine Learning, pp. 4848–4856. PMLR (2018)
Xu, J., Tian, Y., Sun, Y., Scutari, G.: Distributed algorithms for composite optimization: Unified framework and convergence analysis. arXiv e-prints pp. arXiv-2002 (2020)
Ye, H., Luo, L., Zhou, Z., Zhang, T.: Multi-consensus decentralized accelerated gradient descent. arXiv preprint arXiv:2005.00797 (2020)
Yuan, K., Ling, Q., Yin, W.: On the convergence of decentralized gradient descent. SIAM J. Optim. 26(3), 1835–1854 (2016)
Zinkevich, M., Weimer, M., Smola, A.J., Li, L.: Parallelized stochastic gradient descent. In: NIPS, vol. 4, p. 4. Citeseer (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Trimbach, E., Rogozin, A. (2021). An Acceleration of Decentralized SGD Under General Assumptions with Low Stochastic Noise. In: Strekalovsky, A., Kochetov, Y., Gruzdeva, T., Orlov, A. (eds) Mathematical Optimization Theory and Operations Research: Recent Trends. MOTOR 2021. Communications in Computer and Information Science, vol 1476. Springer, Cham. https://doi.org/10.1007/978-3-030-86433-0_8
Download citation
DOI: https://doi.org/10.1007/978-3-030-86433-0_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86432-3
Online ISBN: 978-3-030-86433-0
eBook Packages: Computer ScienceComputer Science (R0)