Multi-Layer Sparse Coding: The Holistic Way

DOI10.1137/18M1183352zbMath1499.65153arXiv1804.09788OpenAlexW2963047604WikidataQ128417206 ScholiaQ128417206MaRDI QIDQ5025774

Aviad Aberdam, Jeremias Sulam, Michael Elad

Publication date: 3 February 2022

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1804.09788

zbMATH Keywords

neural networkssparse codingsparse representationsanalysis and synthesis priorsmulti-layer representations

Mathematics Subject Classification ID

Computational methods for sparse matrices (65F50) Image analysis in multivariate analysis (62H35) Numerical solutions to overdetermined systems, pseudoinverses (65F20) Ill-posedness and regularization problems in numerical linear algebra (65F22) Iterative numerical methods for linear systems (65F10) Approximation algorithms (68W25) Linear operators and ill-posed problems, regularization (47A52) Neural nets and related approaches to inference from stochastic processes (62M45)

Related Items (4)

Prevalence of neural collapse during the terminal phase of deep learning training ⋮ \(\ell^1\)-analysis minimization and generalized (co-)sparsity: when does recovery succeed? ⋮ Adversarial noise attacks of deep learning architectures: stability analysis via sparse-modeled signals ⋮ Regularization by architecture: a deep prior approach for inverse problems

Uses Software

• PDCO
• CoSaMP

Cites Work

• Unnamed Item
• Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers
• Proportional response dynamics in the Fisher market
• Compressed sensing with coherent and redundant dictionaries
• The restricted isometry property and its implications for compressed sensing
• CoSaMP: Iterative signal recovery from incomplete and inaccurate samples
• The cosparse analysis model and algorithms
• Greedy-like algorithms for the cosparse analysis model
• The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
• Atomic Decomposition by Basis Pursuit
• Sparse Modeling for Image and Vision Processing
• Performance Guarantees of the Thresholding Algorithm for the Cosparse Analysis Model
• Penalized and Constrained Optimization: An Application to High-Dimensional Website Advertising
• Sparse representations in unions of bases
• Greed is Good: Algorithmic Results for Sparse Approximation
• Just relax: convex programming methods for identifying sparse signals in noise
• Sparse and Redundant Representations
• From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images
• A Theory for Sampling Signals From a Union of Subspaces
• Coherence-Based Performance Guarantees for Estimating a Sparse Vector Under Random Noise
• Analysis K-SVD: A Dictionary-Learning Algorithm for the Analysis Sparse Model
• Learning Sparsifying Transforms
• Working Locally Thinking Globally: Theoretical Guarantees for Convolutional Sparse Coding
• Multi-Layer Convolutional Sparse Modeling: Pursuit and Dictionary Learning
• Subspace Pursuit for Compressive Sensing Signal Reconstruction
• Analysis versus synthesis in signal priors
• Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization
• Optimally sparse representation in general (nonorthogonal) dictionaries via ℓ ¹ minimization

This page was built for publication: Multi-Layer Sparse Coding: The Holistic Way