Oracle inequalities and optimal inference under group sparsity
Open accessThe DOI matches its Crossref record; 36 of the 36 references it cites were found in a registry or an index.
Abstract
We consider the problem of estimating a sparse linear regression vector β* under a Gaussian noise model, for the purpose of both prediction and model selection. We assume that prior knowledge is available on the sparsity pattern, namely the set of variables is partitioned into prescribed groups, only few of which are relevant in the estimation process. This group sparsity assumption suggests us to consider the Group Lasso method as a means to estimate β*. We establish oracle inequalities for the prediction and ℓ2 estimation errors of this estimator. These bounds hold under a restricted eigenvalue condition on the design matrix. Under a stronger condition, we derive bounds for the estimation error for mixed (2, p)-norms with 1 ≤ p ≤ ∞. When p=∞, this result implies that a thresholded version of the Group Lasso estimator selects the sparsity pattern of β* with high probability. Next, we prove that the rate of convergence of our upper bounds is optimal in a minimax sense, up to a logarithmic factor, for all estimators over a class of group sparse vectors. Furthermore, we establish lower bounds for the prediction and ℓ2 estimation errors of the usual Lasso estimator. Using this result, we demonstrate that the Group Lasso can achieve an improvement in the prediction and estimation errors as compared to the Lasso. An important application of our results is provided by the problem of estimating multiple regression equations simultaneously or multi-task learning. In this case, we obtain refinements of the results in [In Proc. of the 22nd Annual Conference on Learning Theory (COLT) (2009)], which allow us to establish a quantitative advantage of the Group Lasso over the usual Lasso in the multi-task setting. Finally, within the same setting, we show how our results can be extended to more general noise distributions, of which we only require the fourth moment to be finite. To obtain this extension, we establish a new maximal moment inequality, which may be of independent interest.
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
An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation Bias
Zellner, A. (1962) · Journal of the American Statistical Association · doi:10.1080/01621459.1962.10480664 · Cited by 8,401
Hsiao, C. (2014) · doi:10.1017/cbo9781139839327 · Cited by 7,869
Hsiao, C. (2022) · doi:10.1017/9781009057745 · Cited by 7,869
Hsiao, C. (2003) · doi:10.1017/cbo9780511754203 · Cited by 7,869
Model Selection and Estimation in Regression with Grouped Variables
Yuan, M., & Lin, Y. (2006) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2005.00532.x · Cited by 7,261 · cited 1×
“This structured sparsity assumption suggests us to consider the Group Lasso method [39] as a mean to estimate β∗ [see (2.”
MethodWeak Convergence and Empirical Processes
van der Vaart, A. W., & Wellner, J. A. (1996) · Springer Series in Statistics · doi:10.1007/978-1-4757-2545-2 · Cited by 5,014
Introduction to Nonparametric Estimation
Tsybakov, A. B. (2009) · Springer Series in Statistics · doi:10.1007/b13794 · Cited by 2,486
Self-citationSimultaneous analysis of Lasso and Dantzig selector
Bickel, P. J., Ritov, Y., & Tsybakov, A. B. (2009) · The Annals of Statistics · doi:10.1214/08-aos620 · Cited by 2,396
Self-citationStable recovery of sparse overcomplete representations in the presence of noise
Donoho, D. L., Elad, M., & Temlyakov, V. N. (2006) · IEEE Transactions on Information Theory · doi:10.1109/tit.2005.860430 · Cited by 2,183
The Dantzig selector: Statistical estimation when p is much larger than n
Candes, E., & Tao, T. (2007) · The Annals of Statistics · doi:10.1214/009053606000001523 · Cited by 1,747
The Group Lasso for Logistic Regression
Meier, L., Van De Geer, S., & Bühlmann, P. (2008) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2007.00627.x · Cited by 1,674
Self-citationConvex multi-task feature learning
Argyriou, A., Evgeniou, T., & Pontil, M. (2008) · Machine Learning · doi:10.1007/s10994-007-5040-8 · Cited by 1,528
Self-citationWainwright, M. J. (2009) · IEEE Transactions on Information Theory · doi:10.1109/tit.2009.2016018 · Cited by 1,198
Convex Analysis and Nonlinear Optimization
Borwein, J. M., & Lewis, A. S. (2006) · CMS books in mathematics · doi:10.1007/978-0-387-31256-9 · Cited by 780
Consistency of the group Lasso and multiple kernel learning
Bach, F. (2007) · arXiv (Cornell University) · doi:10.48550/arxiv.0707.3390 · Cited by 669
The composite absolute penalties family for grouped and hierarchical variable selection
Zhao, P., Rocha, G., & Yu, B. (2009) · The Annals of Statistics · doi:10.1214/07-aos584 · Cited by 631
Variable selection in nonparametric additive models
Huang, J., Horowitz, J. L., & Wei, F. (2010) · The Annals of Statistics · doi:10.1214/09-aos781 · Cited by 537
Ravikumar, P., Lafferty, J., Liu, H., & Wasserman, L. (2009) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2009.00718.x · Cited by 535
Lenk, P. J., DeSarbo, W. S., Green, P. E., & Young, M. R. (1996) · Marketing Science · doi:10.1287/mksc.15.2.173 · Cited by 532
Huang, J., & Zhang, T. (2010) · The Annals of Statistics · doi:10.1214/09-aos778 · Cited by 521
The sparsity and bias of the Lasso selection in high-dimensional linear regression
Zhang, C.-H., & Huang, J. (2008) · The Annals of Statistics · doi:10.1214/07-aos520 · Cited by 506
High-dimensional additive modeling
Meier, L., van de Geer, S., & Bühlmann, P. (2009) · The Annals of Statistics · doi:10.1214/09-aos692 · Cited by 458
Self-citationHigh-dimensional generalized linear models and the lasso
van de Geer, S. A. (2008) · The Annals of Statistics · doi:10.1214/009053607000000929 · Cited by 414
Self-citationSparsity oracle inequalities for the Lasso
Bunea, F., Tsybakov, A. B., & Wegkamp, M. H. (2007) · HAL · Cited by 389
Self-citationSupport union recovery in high-dimensional multivariate regression
Obozinski, G., Wainwright, M. J., & Jordan, M. I. (2011) · The Annals of Statistics · doi:10.1214/09-aos776 · Cited by 350
Seemingly unrelated regression equations models : estimation and inference
Srivastava, V. K., & Giles, D. E. A. (1987) · Cited by 319
Aggregation for Gaussian regressionPreprint
Bunea, F., Tsybakov, A. B., & Wegkamp, M. H. (2016) · Cited by 296
Self-citationSparsity oracle inequalities for the Lasso
Bunea, F., Tsybakov, A., & Wegkamp, M. (2007) · Electronic Journal of Statistics · doi:10.1214/07-ejs008 · Cited by 253
Self-citationSup-norm convergence rate and sign concentration property of Lasso and Dantzig estimators
Lounici, K. (2008) · Electronic Journal of Statistics · doi:10.1214/08-ejs177 · Cited by 209
Self-citationOracle inequalities for inverse problems
Cavalier, L., Golubev, G. K., Picard, D., et al. (2002) · The Annals of Statistics · doi:10.1214/aos/1028674843 · Cited by 189
Taking Advantage of Sparsity in Multi-Task LearningPreprint
Lounici, K., Pontil, M., Tsybakov, A. B., & van de Geer, S. (2009) · arXiv (Cornell University) · doi:10.48550/arxiv.0903.1468 · Cited by 182
Self-citationAggregation for Gaussian regression
Bunea, F., Tsybakov, A. B., & Wegkamp, M. H. (2007) · The Annals of Statistics · doi:10.1214/009053606000001587 · Cited by 179
Self-citationOn the asymptotic properties of the group lasso estimator for linear models
Nardi, Y., & Rinaldo, A. (2008) · Electronic Journal of Statistics · doi:10.1214/08-ejs200 · Cited by 173
Sup-norm convergence rate and sign concentration property of Lasso and Dantzig estimatorsPreprint
Lounici, K. (2008) · HAL · Cited by 158
Self-citationA Convex Optimization Approach to Modeling Consumer Heterogeneity in Conjoint Estimation
Evgeniou, T., Pontil, M., & Toubia, O. (2007) · Marketing Science · doi:10.1287/mksc.1070.0291 · Cited by 150
Self-citationSome theoretical results on the Grouped Variables Lasso
Chesneau, C., & Hebiri, M. (2008) · Mathematical Methods of Statistics · doi:10.3103/s1066530708040030 · Cited by 57
Cited by
High-Dimensional Statistics: A Non-Asymptotic Viewpoint
Wainwright, M. J. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108627771 · Cited by 882
Least Squares After Model Selection in High-Dimensional Sparse Models
Belloni, A., & Chernozhukov, V. (2009) · SSRN Electronic Journal · doi:10.2139/ssrn.1582594 · Cited by 683
Estimation of high-dimensional low-rank matrices
Rohde, A., & Tsybakov, A. B. (2011) · The Annals of Statistics · doi:10.1214/10-aos860 · Cited by 330
Self-citationRobust inference on average treatment effects with possibly more covariates than observations
Farrell, M. H. (2015) · Journal of Econometrics · doi:10.1016/j.jeconom.2015.06.017 · Cited by 320
Structured Sparsity through Convex Optimization
Bach, F., Jenatton, R., Mairal, J., & Obozinski, G. (2012) · Statistical Science · doi:10.1214/12-sts394 · Cited by 315
Metadata
- Authors
- Lounici, K., Pontil, M., van de Geer, S., & Tsybakov, A. B.
- Venue
- The Annals of Statistics
- Publisher
- Institute of Mathematical Statistics
- Published
- 2011-08-01
- Volume
- 39
- Issue
- 4
- Pages
- 2164-2204
- ISSN
- 0090-5364 (ISSN-L)
- OpenAlex
- W1981657694
- Cited by
- 341
- References
- 36
- FWCI
- 18.64
- 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.