Performance analysis of matrix-free conjugate gradient kernels using SYCL.
Igor A. BarattaChris N. RichardsonGarth N. WellsPublished in: IWOCL (2022)
Keyphrases
- conjugate gradient
- reproducing kernel hilbert space
- positive definite
- kernel machines
- training algorithm
- convergence rate
- levenberg marquardt
- kernel function
- maximum likelihood estimation
- faster convergence
- support vector
- distance measure
- multiple kernel learning
- learning theory
- neural network
- euclidean space
- singular value decomposition
- loss function
- real valued
- learning problems
- kernel methods
- data dependent
- kernel matrix
- low rank
- domain adaptation
- bundle adjustment
- back propagation
- least squares
- special case
- genetic algorithm