Pairwise Conditional Gradients without Swap Steps and Sparser Kernel Herding.
Kazuma TsujiKen'ichiro TanakaSebastian PokuttaPublished in: ICML (2022)
Keyphrases
- pairwise
- similarity function
- loss function
- kernel function
- similarity measure
- markov random field
- graph cuts
- multi class
- higher order
- kernel methods
- pairwise interactions
- kernel matrix
- metric learning
- graph matching
- support vector
- semi supervised
- feature space
- kernel learning
- positive definite
- spectral clustering
- high order
- kernel pca
- kernel machines
- image segmentation