Divide and Conquer the Embedding Space for Metric Learning.
Artsiom SanakoyeuVadim TschernezkiUta BüchlerBjörn OmmerPublished in: CoRR (2019)
Keyphrases
- metric learning
- embedding space
- dimensionality reduction
- kernel matrix
- manifold structure
- graph embedding
- manifold learning
- low dimensional
- distance metric
- semi supervised
- high dimensional
- feature space
- input space
- data points
- euclidean space
- euclidean distance
- geometric structure
- geodesic distance
- pairwise
- nonlinear dimensionality reduction
- distance function
- high dimensional data
- pattern recognition
- principal component analysis
- data representation
- dimensionality reduction methods
- learning tasks
- locally linear embedding
- unsupervised learning
- positive semidefinite
- linear discriminant analysis
- semidefinite programming
- multi task
- feature extraction
- feature selection
- pairwise constraints
- sparse representation
- dimension reduction
- machine learning