A maximum principle argument for the uniform convergence of graph Laplacian regressors.
Nicolás García TrillosRyan MurrayPublished in: CoRR (2019)
Keyphrases
- uniform convergence
- learning rate
- reproducing kernel hilbert space
- euclidean space
- vc dimension
- spectral clustering
- real valued
- random walk
- generalization error
- kernel machines
- learning algorithm
- sample complexity
- model selection
- high dimensional data
- semi supervised
- data points
- riemannian manifolds
- probabilistic model