Homology-Preserving Dimensionality Reduction via Manifold Landmarking and Tearing.
Lin YanYaodong ZhaoPaul RosenCarlos ScheideggerBei WangPublished in: CoRR (2018)
Keyphrases
- dimensionality reduction
- manifold learning
- low dimensional
- lower dimensional
- nonlinear dimensionality reduction
- diffusion maps
- high dimensional
- manifold structure
- locally linear embedding
- graph embedding
- high dimensional data
- principal component analysis
- feature space
- digital images
- laplacian eigenmaps
- nonlinear manifold
- pattern recognition
- subspace learning
- data points
- feature extraction
- data representation
- dimensionality reduction methods
- input space
- pattern recognition and machine learning
- linear discriminant analysis
- high dimensionality
- underlying manifold
- structure preserving
- feature selection
- principal components
- linear dimensionality reduction
- data sets
- locality preserving projections
- euclidean space
- euclidean distance
- metric learning
- random projections
- kernel pca
- dimension reduction
- embedding space
- nearest neighbor
- graph laplacian
- intrinsic dimensionality
- support vector
- neural network
- machine learning
- semi supervised
- shortest path
- vector space
- principal components analysis
- geodesic distance