Minimizing finite sums with the stochastic average gradient
Open accessThe DOI matches its Crossref record; 56 of the 56 references it cites were found in a registry or an index.
Abstract
We analyze the stochastic average gradient (SAG) method for optimizing the sum of a finite number of smooth convex functions. Like stochastic gradient (SG) methods, the SAG method’s iteration cost is independent of the number of terms in the sum. However, by incorporating a memory of previous gradient values the SAG method achieves a faster convergence rate than black-box SG methods. The convergence rate is improved from O(1/k)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O(1/\sqrt{k})$$\end{document} to O(1 / k) in general, and when the sum is strongly-convex the convergence rate is improved from the sub-linear O(1 / k) to a linear convergence rate of the form O(ρk)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O(\rho ^k)$$\end{document} for ρ<1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho < 1$$\end{document}. Further, in many cases the convergence rate of the new method is also faster than black-box deterministic gradient methods, in terms of the number of gradient evaluations. This extends our earlier work Le Roux et al. (Adv Neural Inf Process Syst, 2012), which only lead to a faster rate for well-conditioned strongly-convex problems. Numerical experiments indicate that the new algorithm often dramatically outperforms existing SG and deterministic gradient methods, and that the performance may be further improved through the use of non-uniform sampling strategies.
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
UCI Machine Learning Repository
Asuncion, A. (2007) · Medical Entomology and Zoology · Cited by 25,373
(2021) · A First Course in Statistical Programming with R · doi:10.1017/9781108993456.011 · Cited by 13,724
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
A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging
Chambolle, A., & Pock, T. (2011) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-010-0251-1 · Cited by 4,336
Monte Carlo Statistical Methods
Robert, C. P., & Casella, G. (2004) · Springer Texts in Statistics · doi:10.1007/978-1-4757-4145-2 · Cited by 4,271
Introductory Lectures on Convex Optimization: A Basic Course
Nesterov, Y. E. (2014) · Medical Entomology and Zoology · Cited by 3,557
Introductory Lectures on Convex Optimization
Nesterov, Y. (2004) · Applied Optimization · doi:10.1007/978-1-4419-8853-9 · Cited by 2,698
RCV1: A New Benchmark Collection for Text Categorization Research
Lewis, D., Yang, Y., Rose, T. G., & Li, F. (2004) · Goldsmiths (University of London) · doi:10.5555/1005332.1005345 · Cited by 2,503
Smooth minimization of non-smooth functions
Nesterov, Y. (2005) · Mathematical Programming · doi:10.1007/s10107-004-0552-5 · Cited by 2,455
Robust Stochastic Approximation Approach to Stochastic Programming
Nemirovski, A., Juditsky, A., Lan, G., & Shapiro, A. (2009) · SIAM Journal on Optimization · doi:10.1137/070704277 · Cited by 1,945
Accelerating Stochastic Gradient Descent using Predictive Variance Reduction
Johnson, R., & Zhang, T. (2013) · Cited by 1,889
Acceleration of Stochastic Approximation by Averaging
Polyak, B. T., & Juditsky, A. B. (1992) · SIAM Journal on Control and Optimization · doi:10.1137/0330046 · Cited by 1,827 · cited 2×
“With a suitable choice of step-sizes, this gives the same asymptotic efficiency as Newton-like secondorder SGmethods and also leads to increased robustness of the convergence rate to the exact sequence of step sizes [2,47].”
BackgroundMethodPegasos: primal estimated sub-gradient solver for SVM
Shalev-Shwartz, S., Singer, Y., Srebro, N., & Cotter, A. (2011) · Mathematical Programming · doi:10.1007/s10107-010-0420-4 · Cited by 1,477
Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems
Nesterov, Y. (2012) · SIAM Journal on Optimization · doi:10.1137/100802001 · Cited by 1,187
Gradient methods for minimizing composite objective functionPreprint
Nesterov, Y. E. (2007) · RePEc: Research Papers in Economics · Cited by 1,096
A Randomized Kaczmarz Algorithm with Exponential Convergence
Strohmer, T., & Vershynin, R. (2009) · Journal of Fourier Analysis and Applications · doi:10.1007/s00041-008-9030-4 · Cited by 903
A method for unconstrained convex minimization problem with the rate of convergence o(1/k^2)
Nesterov, Y. E. (1983) · Medical Entomology and Zoology · Cited by 865
Primal-dual subgradient methods for convex problems
Nesterov, Y. (2009) · Mathematical Programming · doi:10.1007/s10107-007-0149-x · Cited by 835
SAGA: A Fast Incremental Gradient Method With Support for Non-Strongly Convex Composite Objectives
Defazio, A., Bach, F., & Lacoste-Julien, S. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1407.0202 · Cited by 834
Self-citationDeep learning via Hessian-free optimization
Martens, J. (2010) · International Conference on Machine Learning · Cited by 732
Stochastic dual coordinate ascent methods for regularized lossPreprint
Shalev‐Shwartz, S., & Zhang, T. (2013) · Cited by 637
A Proximal Stochastic Gradient Method with Progressive Variance Reduction
Xiao, L., & Zhang, T. (2014) · SIAM Journal on Optimization · doi:10.1137/140961791 · Cited by 592
Dual Averaging Method for Regularized Stochastic Learning and Online Optimization
Xiao, L. (2009) · Cited by 574
Making Gradient Descent Optimal for Strongly Convex Stochastic Optimization
Rakhlin, A., Shamir, O., & Sridharan, K. (2011) · arXiv (Cornell University) · doi:10.48550/arxiv.1109.5647 · Cited by 492
Non-Asymptotic Analysis of Stochastic Approximation Algorithms for Machine Learning
Bach, F., & Moulines, É. (2011) · Cited by 488
Self-citationAccelerated Stochastic Approximation
Kesten, H. (1958) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177706705 · Cited by 433
Stochastic gradient descent, weighted sampling, and the randomized Kaczmarz algorithm
Needell, D., Srebro, N., & Ward, R. (2016) · Mathematical Programming · doi:10.1007/s10107-015-0864-7 · Cited by 432
Accelerated proximal stochastic dual coordinate ascent for regularized loss minimization
Shalev-Shwartz, S., & Zhang, T. (2016) · Mathematical Programming · doi:10.1007/s10107-014-0839-0 · Cited by 360
Bottou, L., & Cun, Y. L. (2003) · Cited by 359
SGD-QN: Careful Quasi-Newton Stochastic Gradient Descent
Bordes, A., Bottou, L., & Gallinari, P. (2009) · HAL · Cited by 324
Hybrid deterministic-stochastic methods for data fitting
Friedlander, M. P., & Schmidt, M. (2011) · Cited by 314
Self-citationA New Class of Incremental Gradient Methods for Least Squares Problems
Bertsekas, D. P. (1997) · SIAM Journal on Optimization · doi:10.1137/s1052623495287022 · Cited by 312
A Convergent Incremental Gradient Method with a Constant Step Size
Blatt, D., Hero, A. O., & Gauchman, H. (2007) · SIAM Journal on Optimization · doi:10.1137/040615961 · Cited by 306
Hybrid Deterministic-Stochastic Methods for Data Fitting
Friedlander, M. P., & Schmidt, M. (2012) · SIAM Journal on Scientific Computing · doi:10.1137/110830629 · Cited by 254
Self-citationInformation-Theoretic Lower Bounds on the Oracle Complexity of Stochastic Convex Optimization
Agarwal, A., Bartlett, P. L., Ravikumar, P., & Wainwright, M. J. (2012) · IEEE Transactions on Information Theory · doi:10.1109/tit.2011.2182178 · Cited by 225
Large-scale sparse logistic regression
Liu, J., Chen, J., & Ye, J. (2009) · Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1557019.1557082 · Cited by 191
Convergence Rate of Incremental Subgradient Algorithms
Nedić, A., & Bertsekas, D. (2001) · Applied Optimization · doi:10.1007/978-1-4757-6594-6_11 · Cited by 190
A scalable modular convex solver for regularized risk minimization
Teo, C. H., Smola, A., Vishwanathan, S. V. N., & Le, Q. V. (2007) · Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1281192.1281270 · Cited by 171
Fast Convergence of Stochastic Gradient Descent under a Strong Growth ConditionPreprint
Schmidt, M., & Roux, N. L. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1308.6370 · Cited by 171
Self-citationExponentiated Gradient Algorithms for Conditional Random Fields and Max-Margin Markov Networks
Collins, M., Globerson, A., Koo, T., Carreras, X., & Bartlett, P. L. (2008) · Journal of Machine Learning Research · doi:10.5555/1390681.1442791 · Cited by 165
Lin, Q., Lu, Z., & Xiao, L. (2015) · SIAM Journal on Optimization · doi:10.1137/141000270 · Cited by 163
An Incremental Gradient(-Projection) Method with Momentum Term and Adaptive Stepsize Rule
Tseng, P. (1998) · SIAM Journal on Optimization · doi:10.1137/s1052623495294797 · Cited by 160
Accelerated Gradient Methods for Stochastic Optimization and Online Learning
Hu, C., Pan, W., & Kwok, J. T.-Y. (2009) · Cited by 140
Finito: A Faster, Permutable Incremental Gradient Method for Big Data ProblemsPreprint
Defazio, A., Caetano, T. S., & Domke, J. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1407.2710 · Cited by 126
Block-Coordinate Frank-Wolfe Optimization for Structural SVMs
Lacoste-Julien, S., Jaggi, M., Schmidt, M., & Pletscher, P. (2013) · Cited by 115
Self-citationHazan, E., & Kale, S. (2011) · Cited by 112
Linear Convergence with Condition Number Independent Access of Full Gradients
Zhang, L., Mahdavi, M., & Jin, R. (2013) · Cited by 99
Spam Corpus Creation for TREC.
Cormack, G. V., & Lynam, T. R. (2005) · Cited by 84
Non-Uniform Stochastic Average Gradient Method for Training Conditional Random FieldsPreprint
Schmidt, M., Babanezhad, R., Ahmed, M. O., Defazio, A., Clifton, A., & Sarkar, A. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1504.04406 · Cited by 78
Self-citationStochastic Optimization with Importance SamplingPreprint
Zhao, P., & Zhang, T. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1401.2753 · Cited by 77
Fast large-scale optimization by unifying stochastic gradient and quasi-Newton methods
Sohl‐Dickstein, J., Poole, B., & Ganguli, S. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1311.2115 · Cited by 74
Accelerated Stochastic Approximation
Delyon, B., & Juditsky, A. (1993) · SIAM Journal on Optimization · doi:10.1137/0803045 · Cited by 61
Randomized Dual Coordinate Ascent with Arbitrary SamplingPreprint
Qu, Z., Richtárik, P., & Zhang, T. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1411.5873 · Cited by 49
MixedGrad: An O(1/T) Convergence Rate Algorithm for Stochastic Smooth OptimizationPreprint
Mahdavi, M., & Jin, R. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1307.7192 · Cited by 10
Carbonetto, P. S. (2009) · cIRcle (University of British Columbia) · doi:10.14288/1.0051537 · Cited by 5
Cited by
Optimization Methods for Large-Scale Machine Learning
Bottou, L., Curtis, F. E., & Nocedal, J. (2018) · SIAM Review · doi:10.1137/16m1080173 · Cited by 3,128
SCAFFOLD: Stochastic Controlled Averaging for Federated Learning
Karimireddy, S. P., Kale, S., Mohri, M., Reddi, S. J., Stich, S. U., & Suresh, A. T. (2019) · arXiv · doi:10.48550/arxiv.1910.06378 · Cited by 1,734
“In optimization, they were used for finite-sum minimization by SVRG (Johnson and Zhang, 2013; Zhang et al., 2013) and then in SAGA (Defazio et al., 2014) to simplify the linearly convergent method SAG (Schmidt et al., 2017).”
MethodInfluentialFederated Optimization: Distributed Machine Learning for On-Device Intelligence
Konečný, J., McMahan, H. B., Ramage, D., & Richtárik, P. (2016) · arXiv · doi:10.48550/arxiv.1610.02527 · Cited by 1,563
Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithms
Zhang, K., Yang, Z., & Başar, T. (2021) · Studies in Systems, Decision and Control · doi:10.1007/978-3-030-60990-0_12 · Cited by 1,264
Federated Learning on Non-IID Data Silos: An Experimental Study
Li, Q., Diao, Y., Chen, Q., & He, B. (2022) · 2022 IEEE 38th International Conference on Data Engineering (ICDE) · doi:10.1109/icde53745.2022.00077 · Cited by 1,132
Metadata
- Authors
- Schmidt, M., Le Roux, N., & Bach, F.
- Venue
- Mathematical Programming
- Published
- 2016-06-14
- Volume
- 162
- Issue
- 1-2
- Pages
- 83-112
- ISSN
- 0025-5610 (ISSN-L)
- 1436-4646
- OpenAlex
- W1791038712
- Cited by
- 922
- References
- 56
- FWCI
- 88.25
- Type
- journal article
- 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
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.