Sparse group Lasso and high dimensional multinomial classification
From MaRDI portal
Abstract: The sparse group lasso optimization problem is solved using a coordinate gradient descent algorithm. The algorithm is applicable to a broad class of convex loss functions. Convergence of the algorithm is established, and the algorithm is used to investigate the performance of the multinomial sparse group lasso classifier. On three different real data examples the multinomial group lasso clearly outperforms multinomial lasso in terms of achieved classification error rate and in terms of including fewer features for the classification. The run-time of our sparse group lasso implementation is of the same order of magnitude as the multinomial lasso algorithm implemented in the R package glmnet. Our implementation scales well with the problem size. One of the high dimensional examples considered is a 50 class classification problem with 10k features, which amounts to estimating 500k parameters. The implementation is available as the R package msgl.
Recommendations
- Subspace quadratic regularization method for group sparse multinomial logistic regression
- The Group Lasso for Logistic Regression
- A fast unified algorithm for solving group-lasso penalize learning problems
- Block coordinate descent algorithms for large-scale sparse multiclass classification
- Sparse optimization for nonconvex group penalized estimation
Cites work
- scientific article; zbMATH DE number 2121575 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- A coordinate gradient descent method for nonsmooth separable minimization
- Convergence of a block coordinate descent method for nondifferentiable minimization
- Input selection and shrinkage in multiresponse linear regression
- Multiclass sparse logistic regression for classification of multiple cancer types using gene expression data
- RcppArmadillo: accelerating R with high-performance C++ linear algebra
- The Group Lasso for Logistic Regression
Cited in
(31)- High-Dimensional Gaussian Graphical Regression Models with Covariates
- Two-layer feature reduction for sparse-group Lasso via decomposition of convex sets
- Feature selection for classification models via bilevel optimization
- Informative gene selection for microarray classification via adaptive elastic net with conditional mutual information
- On the proximal Landweber Newton method for a class of nonsmooth convex problems
- Sparse group lasso for multiclass functional logistic regression models
- Posterior contraction in group sparse logit models for categorical responses
- Multiresolution categorical regression for interpretable cell-type annotation
- Penalized logistic regression with prior information for microarray gene expression classification
- Simultaneous variable and factor selection via sparse group lasso in factor analysis
- AdaBoost Semiparametric Model Averaging Prediction for Multiple Categories
- Using the “Hidden” genome to improve classification of cancer types
- Variable selection for naïve Bayes classification
- On the linear convergence of a proximal gradient method for a class of nonsmooth convex minimization problems
- Structured sparsity: discrete and convex approaches
- Subspace quadratic regularization method for group sparse multinomial logistic regression
- Benchmarking penalized regression methods in machine learning for single cell RNA sequencing data
- Block coordinate descent algorithms for large-scale sparse multiclass classification
- Robust and sparse multigroup classification by the optimal scoring approach
- A Unified Approach to Sparse Tweedie Modeling of Multisource Insurance Claim Data
- Stochastic DCA for minimizing a large sum of DC functions with application to multi-class logistic regression
- An Iterative Sparse-Group Lasso
- Multiclass-penalized logistic regression
- A modified proximal gradient method for a family of nonsmooth convex optimization problems
- Robust inference on average treatment effects with possibly more covariates than observations
- MCEN: a method of simultaneous variable selection and clustering for high-dimensional multinomial regression
- A Likelihood-Based Approach for Multivariate Categorical Response Regression in High Dimensions
- Variable selection in general multinomial logit models
- Non-convex penalized multitask regression using data depth-based penalties
- Automatic Response Category Combination in Multinomial Logistic Regression
- A modified multinomial baseline logit model with logit functions having different covariates
Describes a project that uses
Uses Software
This page was built for publication: Sparse group Lasso and high dimensional multinomial classification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1621358)
