Differentiable Kernels in Generalized Matrix Learning Vector Quantization.
Marika KästnerDavid NebelMartin RiedelMichael BiehlThomas VillmannPublished in: ICMLA (1) (2012)
Keyphrases
- learning vector quantization
- positive semidefinite
- positive definite
- neural network
- bankruptcy prediction
- self organizing maps
- matrix valued
- eigenvalue problems
- kernel matrices
- semidefinite programming
- linear combination
- kernel function
- semidefinite
- objective function
- eigenvalue decomposition
- rough approximations
- linear algebra
- multiple kernel learning
- kernel matrix
- support vector
- metric learning
- low rank
- gram matrix
- kernel methods
- image data
- positive semi definite
- moore penrose