Objective functions for optimization.
In shark, the learning problem is phrased as an objective function which is then optimized using Algorithms to find a local minimum ofObjective functions. . This allows to test and develop algorithms using Benchmark functions independent of the problem to solve.
Collaboration diagram for Objective functions:
Classes | |
| class | shark::AbstractObjectiveFunction< PointType, ResultT > |
| Super class of all objective functions for optimization and learning. More... | |
| class | shark::CombinedObjectiveFunction< SearchPointType, ResultT > |
| Linear combination of objective functions. More... | |
| class | shark::CrossValidationError< ModelTypeT, LabelTypeT > |
| Cross-validation error for selection of hyper-parameters. More... | |
| class | shark::ErrorFunction< SearchPointType > |
| Objective function for supervised learning. More... | |
| class | shark::EvaluationArchive< PointType, ResultT > |
| Objective function wrapper storing all function evaluations. More... | |
| class | shark::LooError< ModelTypeT, LabelType > |
| Leave-one-out error objective function. More... | |
| class | shark::NegativeLogLikelihood |
| Computes the negative log likelihood of a dataset under a model. More... | |
| class | shark::OneNormRegularizer< SearchPointType > |
| One-norm of the input as an objective function. More... | |
| class | shark::TwoNormRegularizer< SearchPointType > |
| Two-norm of the input as an objective function. More... | |
| class | shark::VariationalAutoencoderError< SearchPointType > |
| Computes the variational autoencoder error function. More... | |
Modules | |
| Constraint Handling | |
| Objects for handling constraints. | |
| Benchmark functions | |
| Kernel Optimization | |
| All kinds of objective functions to optimize kernel functions. | |