Non-linear, Sparse Dimensionality Reduction via Path Lasso Penalized Autoencoders.
Oskar AllerboRebecka JörnstenPublished in: CoRR (2021)
Keyphrases
- dimensionality reduction
- sparse representation
- high dimensional
- variable selection
- group lasso
- random projections
- least squares
- high dimension
- high dimensional data
- regularized regression
- feature selection
- kernel trick
- elastic net
- sparse coding
- low dimensional
- principal component analysis
- denoising
- compressed sensing
- high dimensionality
- input space
- data representation
- feature extraction
- dimensionality reduction methods
- generalized linear models
- kernel learning
- manifold learning
- compressive sensing
- dimension reduction
- linear discriminant analysis
- unsupervised learning
- pattern recognition
- model selection
- nonlinear dimensionality reduction
- singular value decomposition
- pattern recognition and machine learning
- principal components
- basis pursuit
- regression model
- locally linear embedding
- piece wise linear
- shortest path
- feature space
- linear regression
- fisher discriminant analysis
- multiple kernel learning
- sparse data
- structure preserving
- face recognition
- data points
- nearest neighbor
- cross validation
- efficient optimization
- graph embedding
- knn
- kernel methods
- loss function
- maximum likelihood