Combining Wasserstein-1 and Wasserstein-2 proximals: robust manifold learning via well-posed generative flows.
Hyemin GuMarkos A. KatsoulakisLuc Rey-BelletBenjamin J. ZhangPublished in: CoRR (2024)
Keyphrases
- manifold learning
- pointwise
- low dimensional
- nonlinear dimensionality reduction
- diffusion maps
- semi supervised
- high dimensional
- dimensionality reduction
- subspace learning
- manifold structure
- dimension reduction
- laplacian eigenmaps
- high dimensional data
- generative model
- feature extraction
- sparse representation
- feature space
- discriminant projection
- feature vectors
- head pose estimation
- unsupervised learning
- locality preserving
- data points
- manifold learning algorithm
- discriminant embedding