Learning with Structured Sparsity
From MaRDI portal
Abstract: This paper investigates a new learning formulation called structured sparsity, which is a natural extension of the standard sparsity concept in statistical learning and compressive sensing. By allowing arbitrary structures on the feature set, this concept generalizes the group sparsity idea that has become popular in recent years. A general theory is developed for learning with structured sparsity, based on the notion of coding complexity associated with the structure. It is shown that if the coding complexity of the target signal is small, then one can achieve improved performance by using coding complexity regularization methods, which generalize the standard sparse regularization. Moreover, a structured greedy algorithm is proposed to efficiently solve the structured sparsity problem. It is shown that the greedy algorithm approximately solves the coding complexity optimization problem under appropriate conditions. Experiments are included to demonstrate the advantage of structured sparsity over standard sparsity on some real applications.
Recommendations
- Learning the Structure for Structured Sparsity
- Structured sparsity and generalization
- Regularizers for structured sparsity
- Structured sparsity through convex optimization
- Structured sparsity via alternating direction methods
- Structured sparsity: discrete and convex approaches
- Structured sparsity promoting functions
- Sparse low-rank separated representation models for learning from data
- Consistent Sparse Deep Learning: Theory and Computation
Cited in
(44)- Linearized alternating direction method of multipliers for sparse group and fused Lasso models
- Two-level structural sparsity regularization for identifying lattices and defects in noisy images
- Robust visual tracking via consistent low-rank sparse learning
- Structured learning with constrained conditional models
- Sequential approaches for learning datum-wise sparse representations
- Regularizers for structured sparsity
- Theoretical properties of the overlapping groups Lasso
- Traditional and recent approaches in background modeling for foreground detection: an overview
- Beyond covariance: SICE and kernel based visual feature representation
- On change-point estimation under Sobolev sparsity
- Efficient inexact proximal gradient algorithms for structured sparsity-inducing norm
- Cluster sparsity field: an internal hyperspectral imagery prior for reconstruction
- Discrete optimization methods for group model selection in compressed sensing
- Sparse trace norm regularization
- Investigating consumers' store-choice behavior via hierarchical variable selection
- Learning Reductions to Sparse Sets
- Designing Structured Sparse Dictionaries for Sparse Representation Modeling
- Smooth sparse coding via marginal regression for learning sparse representations
- Structured sparsity: discrete and convex approaches
- Grouping strategies and thresholding for high dimensional linear models
- Sparsity constrained estimation in image processing and computer vision
- Learning Sparsifying Transforms
- Learning the Structure for Structured Sparsity
- Bayesian inference for spatio-temporal spike-and-slab priors
- Fast algorithms for structured sparsity (ICALP 2015 invited tutorial)
- ORKA: Object reconstruction using a K-approximation graph
- Structured sparse support vector machine with ordered features
- Overlapping group lasso for high-dimensional generalized linear models
- scientific article; zbMATH DE number 7266294 (Why is no real title available?)
- Decomposition into low-rank plus additive matrices for background/foreground separation: a review for a comparative evaluation with a large-scale dataset
- Dependent relevance determination for smooth and structured sparse regression
- Sparse low-rank separated representation models for learning from data
- Kernelized elastic net regularization: generalization bounds, and sparse recovery
- Minimum description length penalization for group and multi-task sparse learning
- Structured sparsity and generalization
- Deep Learning as Sparsity-Enforcing Algorithms
- Structured sparsity through convex optimization
- Learning sparse linear combinations of basis functions over a finite domain
- Comment on: \(\ell _{1}\)-penalization for mixture regression models
- Structured, sparse regression with application to HIV drug resistance
- An improved tensor regression model via location smoothing
- Foveated compressed sensing
- Convex relaxations of penalties for sparse correlated variables with bounded total variation
- Theoretical guarantees for graph sparse coding
This page was built for publication: Learning with Structured Sparsity
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5396730)
