SpAM: Sparse Additive Models

By Pradeep Ravikumar, Han Liu, John D. Lafferty, Larry A. Wasserman. 2007

204 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Breaking the Curse of Dimensionality with Convex Neural Networks (2014, background): “(b) Generalized additive models: f(x) = ∑d j=1 fj(xj), which are generalizations of the above by summing functions fj : R→ R which may not be affine (Hastie and Tibshirani, 1990; Ravikumar et al., 2008; Bach, 2008a).”
Sparse additive models (2007, methodology): “10. Acknowledgements. This research was supported in part by NSF grant CCF-0625879 and a Siebel Scholarship to PR. A preliminary version of part of this work appears in Ravikumar et al. (2008) .”
High-dimensional additive modeling (2008, methodology): “d by a linear model, substantial improvements are sometimes possible by using a more ﬂexible class of models. Recently, some progress has been made regarding high-dimensional additive model selection [7, 19, 26] and some theoretical results are available [26]. Other approaches are based on wavelets ”
Taking Advantage of Sparsity in Multi-Task Learning (2009, methodology): “[25] focused on the consistent selection, whereas it remained unclear whether there is some improvement in the prediction properties as compared to the usual Lasso.”
Consistency of the group Lasso and multiple kernel learning (2007, unknown): “e conditional expectation of Y given sums of measurable functions of Xj is attained at functions in the RKHS, and not merely measurable functions. Dealing with more general assumptions in the line of Ravikumar et al. (2008) requires to consider consistency for norms weaker than the RKHS norms (Capon”
Bayesian Sparse Group Selection (2016, background): “Another example is sparse additive model, where the nonlinear function of each variable is coded by a group of basis functions (Ravikumar et al. 2008).”
The Lasso with within group structure (2010, background): “Tibshirani [1996]. We show in this paper that β̂ satisfies a sparsity oracle inequality (see Theorem 6.”
Nonparametric sparsity and regularization (2012, background): “o-norm and the ‘1 norm are 0(f) = #fa2f1;:::;dg: kf ak6= 0g and 1(f) = P d a=1 kf ak, respectively. This latter setting, related to multiple kernel learning [8, 6], has been considered for example in [48], see also [36] and references therein. Considering additive models limits the way in which the ”
Identification of Partially Linear Structure in Additive Models with an Application to Gene Expression Prediction from Sequences (2012, background): “For the latter, several works have considered using penalization approaches to select significant components (Ravikumar et al., 2008; Meier, Van de Geer, and Buhlmann, 2009; Xue, 2009; Huang, Horowitz, and Wei, 2010).”
LASSO Isotone for High-Dimensional Additive Isotonic Regression (2010, methodology): “olution paths. Generalisation of the L1 penalisation principle to nonparametric regression can also lead to successful with additive modelling. For example, recent work on this subject includes SpAM [Ravikumar et al. 2007], which describes the application of the grouped LASSO to general smoothers, a”
Minimax-Optimal Rates For Sparse Additive Models Over Kernel Classes Via Convex Programming (2010, methodology): “3 Comparison with other work It is interesting to compare these convergence rates in L(2)(Pn) error with those established in past work [17, 23, 28].”
Deep Neural Networks for High Dimension, Low Sample Size Data (2017, background): “For example, SpAM [Ravikumar et al., 2007] sequentially selects an individual Gradients Selected feature Candidate feature”
Estimation Consistency of the Group Lasso and its Applications (2009, methodology): “The sparse additive models proposed in (Ravikumar et al., 2007) focus more on the consistency analysis of risk and variable selection, which are complementary to our estimation consistency analysis.”
Penalized Estimation in Additive Regression with High-Dimensional Data (2017, methodology): “See, for example, Lin & Zhang (2006), Meier et al. (2009), Ravikumar et al. (2009), Huang et al. (2010), Koltchinskii & Yuan (2010), Raskutti et al. (2012), Suzuki & Sugiyama (2013), Petersen et al. (2016), and Yuan & Zhou (2016).”