Alternating direction multiplier method for matrix l₂,1-norm optimization in multitask feature learning problems
Summary: The joint feature selection problem can be resolved by solving a matrix \(l_{2,1}\)-norm minimization problem. For \(l_{2,1}\)-norm regularization, one of the most fascinating features is that some similar sparsity structures can be employed by multiple predictors. However, the nonsmooth nature of the problem brings great challenges to the problem. In this paper, an alternating direction multiplier method combined with the spectral gradient method is proposed for solving the matrix \(l_{2,1}\)-norm optimization problem involved with multitask feature learning. Numerical experiments show the effectiveness of the proposed algorithm.
- A proximal alternating direction method for \(\ell_{2,1}\)-norm least squares problem in multi-task feature learning
- A unified algorithm for mixed \(l_{2,p}\)-minimizations and its application in feature selection
- Least absolute deviations learning of multiple tasks
- Algorithmic study of \(l_{2,1}\)-norm minimization problems
- scientific article; zbMATH DE number 7312307
- A proximal alternating direction method for \(\ell_{2,1}\)-norm least squares problem in multi-task feature learning
- A regularized alternating direction method of multipliers for a class of nonconvex problems
- Alternating direction algorithms for \(\ell_1\)-problems in compressive sensing
- Consensus-based distributed support vector machines
- Convex multi-task feature learning
- Distributed optimization and statistical learning via the alternating direction method of multipliers
- Efficient online and batch learning using forward backward splitting
- scientific article; zbMATH DE number 5957307 (Why is no real title available?)
- scientific article; zbMATH DE number 3852340 (Why is no real title available?)
- scientific article; zbMATH DE number 51132 (Why is no real title available?)
- scientific article; zbMATH DE number 2076856 (Why is no real title available?)
- Infeasibility detection in the alternating direction method of multipliers for convex optimization
- Learning multiple tasks with kernel methods
- Local linear convergence of the alternating direction method of multipliers on quadratic or linear programs
- On the global and linear convergence of the generalized alternating direction method of multipliers
- Sparse regression using mixed norms
- The Barzilai and Borwein Gradient Method for the Large Scale Unconstrained Minimization Problem
- The Barzilai and Borwein gradient method with nonmonotone line search for nonsmooth convex optimization problems
- The Split Bregman Method for L1-Regularized Problems
- Two-Point Step Size Gradient Methods
- A joint matrix minimization approach for multi-image face recognition
- Least absolute deviations learning of multiple tasks
- A proximal alternating direction method for \(\ell_{2,1}\)-norm least squares problem in multi-task feature learning
- A unified algorithm for mixed \(l_{2,p}\)-minimizations and its application in feature selection
- scientific article; zbMATH DE number 6723396 (Why is no real title available?)
- Algorithmic study of \(l_{2,1}\)-norm minimization problems
This page was built for publication: Alternating direction multiplier method for matrix \(l_{2,1}\)-norm optimization in multitask feature learning problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2007113)
