Sparse and Spurious: Dictionary Learning With Noise and Outliers
From MaRDI portal
Abstract: A popular approach within the signal processing and machine learning communities consists in modelling signals as sparse linear combinations of atoms selected from a learned dictionary. While this paradigm has led to numerous empirical successes in various fields ranging from image to audio processing, there have only been a few theoretical arguments supporting these evidences. In particular, sparse coding, or sparse dictionary learning, relies on a non-convex procedure whose local minima have not been fully analyzed yet. In this paper, we consider a probabilistic model of sparse signals, and show that, with high probability, sparse coding admits a local minimum around the reference dictionary generating the signals. Our study takes into account the case of over-complete dictionaries, noisy signals, and possible outliers, thus extending previous work limited to noiseless settings and/or under-complete dictionaries. The analysis we conduct is non-asymptotic and makes it possible to understand how the key quantities of the problem, such as the coherence or the level of noise, can scale with respect to the dimension of the signals, the number of atoms, the sparsity and the number of observations.
Recommendations
- Sparsity-Driven Laplacian-Regularized Outlier Identification for Dictionary Learning
- Sparsity-Based Poisson Denoising With Dictionary Learning
- Sparsity and nullity: paradigms for analysis dictionary learning
- Double Sparsity: Learning Sparse Dictionaries for Sparse Signal Approximation
- Learning to Sense Sparse Signals: Simultaneous Sensing Matrix and Sparsifying Dictionary Optimization
- Efficient dictionary learning with sparseness-enforcing projections
- Dictionary Learning for Sparse Approximations With the Majorization Method
- Dictionary Learning for L1-Exact Sparse Coding
- Low-rank and sparse dictionary learning
- A max-margin dictionary learning algorithm for sparse representation
Cited in
(19)- Convergence radius and sample complexity of ITKM algorithms for dictionary learning
- Customized dictionary learning for subdatasets with fine granularity
- Fast overcomplete dictionary construction with probabilistic guarantees
- Compressed dictionary learning
- A survey on deep matrix factorizations
- SONIC: social network analysis with influencers and communities
- Learning sparsely used overcomplete dictionaries via alternating minimization
- Generalization Error in Deep Learning
- Local identifiability of \(\ell_1\)-minimization dictionary learning: a sufficient and almost necessary condition
- An improved analysis of the ER-SpUD dictionary learning algorithm
- Unique sharp local minimum in \(\ell_1\)-minimization complete dictionary learning
- Multilinear compressive sensing and an application to convolutional linear networks
- Identifiability of complete dictionary learning
- Provably accurate double-sparse coding
- On the Purity and Entropy of Mixed Gaussian States
- Alternating proximal regularized dictionary learning
- Convergence regions of alternating minimization algorithms for dictionary learning
- Simple alternating minimization provably solves complete dictionary learning
- Data-driven methods for quantitative imaging
This page was built for publication: Sparse and Spurious: Dictionary Learning With Noise and Outliers
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2977139)
