Unified frameworks for high order Newton-Schulz and Richardson iterations: a computationally efficient toolkit for convergence rate improvement.
Alexander StotskyPublished in: J. Appl. Math. Comput. (2019)
Keyphrases
- high order
- convergence rate
- computationally efficient
- stopping criterion
- primal dual
- learning rate
- convergence speed
- higher order
- step size
- low order
- number of iterations required
- global convergence
- gradient method
- pairwise
- numerical stability
- computational efficiency
- markov random field
- fourth order
- tensor analysis
- lower order
- shape representation
- faster convergence rate