Double precision rational approximation algorithms for the standard normal first and second order loss functions.
Steven K. De SchrijverEl Houssaine AghezzafHendrik VanmaelePublished in: Appl. Math. Comput. (2012)
Keyphrases
- approximation algorithms
- loss function
- np hard
- special case
- pairwise
- vertex cover
- squared error
- worst case
- loss minimization
- learning to rank
- support vector
- higher order
- constant factor
- risk minimization
- hinge loss
- approximation ratio
- disjoint paths
- primal dual
- reproducing kernel hilbert space
- polynomial time approximation
- multi class
- convex loss functions
- integer programming
- randomized algorithms
- pairwise constraints
- similarity measure
- precision and recall
- computational complexity
- training data