Algorithm 800: Fortran 77 subroutines for computing the eigenvalues of Hamiltonian matrices. I: the square-reduced method.
Peter BennerRalph ByersEric BarthPublished in: ACM Trans. Math. Softw. (2000)
Keyphrases
- preprocessing
- high accuracy
- improved algorithm
- computational cost
- cost function
- dynamic programming
- optimization algorithm
- computational complexity
- theoretical analysis
- detection algorithm
- segmentation algorithm
- optimization method
- synthetic and real images
- detection method
- experimental evaluation
- k means
- significant improvement
- input data
- objective function
- segmentation method
- energy function
- recognition algorithm
- computationally efficient
- principal components
- sparse matrix
- matching algorithm
- clustering method
- classification method
- simulated annealing
- support vector machine svm
- selection algorithm
- coefficient matrix
- negative matrix factorization
- correlation matrix
- similarity measure
- probabilistic model
- expectation maximization
- tree structure
- covariance matrices
- np hard
- learning algorithm
- matrix representation
- optimal solution
- singular values
- reconstruction method
- search space
- parallel implementation
- edge detection
- em algorithm
- convergence rate
- missing data