A Simple, Pipelined Algorithm for Large, Irregular All-gather Problems.
Jesper Larsson TräffAndreas RipkeChristian SiebertPavan BalajiRajeev ThakurWilliam GroppPublished in: PVM/MPI (2008)
Keyphrases
- learning algorithm
- experimental evaluation
- cost function
- benchmark problems
- detection algorithm
- improved algorithm
- significant improvement
- dynamic programming
- particle swarm optimization
- computational cost
- optimal solution
- clustering method
- parallel implementation
- matching algorithm
- expectation maximization
- high accuracy
- larger problems
- preprocessing
- probabilistic model
- optimization problems
- linear programming
- similarity measure
- recognition algorithm
- computationally efficient
- objective function
- segmentation algorithm
- ant colony optimization
- computational complexity
- linear array
- data sets
- theoretical guarantees
- greedy algorithms
- optimal or near optimal
- combinatorial optimization
- path planning
- energy function
- optimization algorithm
- theoretical analysis
- np complete
- np hard
- data structure
- decision trees