InKS: a programming model to decouple algorithm from optimization in HPC codes.
Ksander EjjaaouaniOlivier AumageJulien BigotMichel MehrenbergerHitoshi MuraiMasahiro NakaoMitsuhisa SatoPublished in: J. Supercomput. (2020)
Keyphrases
- optimization model
- optimization algorithm
- probabilistic model
- theoretical analysis
- cost function
- optimization process
- optimization method
- algorithm employs
- mathematical model
- classification algorithm
- optimization procedure
- objective function
- detection algorithm
- computational complexity
- recognition algorithm
- np hard
- dynamic programming
- preprocessing
- input data
- expectation maximization
- learning algorithm
- em algorithm
- parameter estimation
- optimal solution
- constrained optimization
- estimation algorithm
- tree structure
- search space
- prior information
- high level
- selection algorithm
- energy function
- closed form
- linear programming
- bayesian networks
- bayesian framework
- matching algorithm
- k means
- segmentation algorithm
- simulated annealing
- probability distribution
- decision trees
- multi objective